{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"basic_two_params\",\n  \"description\": \"Basic function with two parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"param1\": {\n        \"type\": \"string\"\n      },\n      \"param2\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"param1\",\n      \"param2\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with hello and world<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\", \"param2\": \"world\"}}", "rejected": "{\"name\": \"basic_two_params\", \"arguments\": {\"param1\": \"hello\" \"param2\": \"world\"}}"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"three_params\",\n  \"description\": \"Function with three parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"number\": {\n        \"type\": \"integer\"\n      },\n      \"flag\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\",\n      \"number\",\n      \"flag\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with test text, number 42, and true flag<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"three_params\", \"arguments\": {\"text\": \"test text\", \"number\": 42, \"flag\": true}}", "rejected": "I will call that function"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"four_params\",\n  \"description\": \"Function with four parameters\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"str1\": {\n        \"type\": \"string\"\n      },\n      \"str2\": {\n        \"type\": \"string\"\n      },\n      \"num\": {\n        \"type\": \"integer\"\n      },\n      \"bool\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"str1\",\n      \"str2\",\n      \"num\",\n      \"bool\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCall with first string, second string, number 10, and false<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"four_params\", \"arguments\": {\"str1\": \"first string\", \"str2\": \"second string\", \"num\": 10, \"bool\": false}}", "rejected": "I will call with those parameters"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Simple text with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Simple text\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Simple text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with punctuation! with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with punctuation!\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with punctuation!"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with numbers 123 with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with numbers 123\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with numbers 123"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with special chars @#$ with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with special chars @#$\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with special chars @#$"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Multi word text string with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Multi word text string\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Multi word text string"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text with hyphen-words with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text with hyphen-words\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text with hyphen-words"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text.with.periods with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text.with.periods\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text.with.periods"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"process_text\",\n  \"description\": \"Process text input\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"input_text\": {\n        \"type\": \"string\"\n      },\n      \"operation\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"input_text\",\n      \"operation\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nProcess this text: Text_with_underscores with analyze operation<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"process_text\", \"arguments\": {\"input_text\": \"Text_with_underscores\", \"operation\": \"analyze\"}}", "rejected": "I will process that text: Text_with_underscores"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nAnalyze sentiment of: The product was excellent<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"The product was excellent\", \"language\": \"en\", \"include_emotions\": true}}", "rejected": "I will analyze the sentiment of that text"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"analyze_sentiment\",\n  \"description\": \"Analyze text sentiment\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"text\": {\n        \"type\": \"string\"\n      },\n      \"language\": {\n        \"type\": \"string\"\n      },\n      \"include_emotions\": {\n        \"type\": \"boolean\"\n      },\n      \"confidence_threshold\": {\n        \"type\": \"number\"\n      }\n    },\n    \"required\": [\n      \"text\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nCheck sentiment for I am frustrated with this service with details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"analyze_sentiment\", \"arguments\": {\"text\": \"I am frustrated with this service\", \"language\": \"en\", \"include_emotions\": true, \"confidence_threshold\": 0.8}}", "rejected": "I will check the sentiment with details"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"units\": {\n        \"type\": \"string\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet 3-day weather for San Francisco in metric units<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"San Francisco\", \"days\": 3, \"units\": \"metric\", \"include_hourly\": false}}", "rejected": "I will get the weather forecast for San Francisco"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"get_weather_forecast\",\n  \"description\": \"Get weather forecast\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"location\": {\n        \"type\": \"string\"\n      },\n      \"days\": {\n        \"type\": \"integer\"\n      },\n      \"include_hourly\": {\n        \"type\": \"boolean\"\n      }\n    },\n    \"required\": [\n      \"location\",\n      \"days\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nGet tomorrow weather for London with hourly details<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"get_weather_forecast\", \"arguments\": {\"location\": \"London\", \"days\": 1, \"include_hourly\": true}}", "rejected": "I will get tomorrow weather for London"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"include_fees\": {\n        \"type\": \"boolean\"\n      },\n      \"precision\": {\n        \"type\": \"integer\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 500 USD to EUR with fees<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 500, \"from_currency\": \"USD\", \"to_currency\": \"EUR\", \"include_fees\": true, \"precision\": 2}}", "rejected": "I will convert that currency for you"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"convert_currency\",\n  \"description\": \"Convert currency amounts\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"amount\": {\n        \"type\": \"number\"\n      },\n      \"from_currency\": {\n        \"type\": \"string\"\n      },\n      \"to_currency\": {\n        \"type\": \"string\"\n      },\n      \"date\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"amount\",\n      \"from_currency\",\n      \"to_currency\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nConvert 250 EUR to CAD using rates from 2023-12-01<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"convert_currency\", \"arguments\": {\"amount\": 250, \"from_currency\": \"EUR\", \"to_currency\": \"CAD\", \"date\": \"2023-12-01\"}}", "rejected": "I will convert using historical rates"}
{"prompt": "<|im_start|>system\nYou are a helpful assistant that calls functions by responding with valid JSON when given a schema. Always respond with JSON function calls only, never prose.<|im_end|>\n\n<schema>\n{\n  \"name\": \"optimize_database_query\",\n  \"description\": \"Optimize database query\",\n  \"parameters\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"sql_query\": {\n        \"type\": \"string\"\n      },\n      \"database_type\": {\n        \"type\": \"string\"\n      },\n      \"performance_target\": {\n        \"type\": \"string\"\n      }\n    },\n    \"required\": [\n      \"sql_query\",\n      \"database_type\"\n    ]\n  }\n}\n</schema>\n\n<|im_start|>user\nOptimize this MySQL query: SELECT name FROM users WHERE active = 1<|im_end|>\n<|im_start|>assistant\n", "chosen": "{\"name\": \"optimize_database_query\", \"arguments\": {\"sql_query\": \"SELECT name FROM users WHERE active = 1\", \"database_type\": \"mysql\", \"performance_target\": \"speed\"}}", "rejected": "I will optimize that MySQL query"}