Dataset Viewer
data_source
stringclasses 3
values | prompt
listlengths 1
1
| ability
stringclasses 2
values | reward_model
dict | extra_info
dict |
|---|---|---|---|---|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t O \t O \t\n _ \t _ \t _ \t P \t\n G \t O \t O \t _ \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100000,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t O \t _ \t _ \t\n G \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t P \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100001,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t _ \t _ \t _ \t\n _ \t _ \t _ \t G \t\n _ \t _ \t _ \t _ \t\n O \t O \t P \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100002,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t P \t G \t _ \t\n O \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100003,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t O \t O \t P \t\n _ \t _ \t _ \t O \t\n _ \t _ \t G \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100004,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t G \t _ \t _ \t\n _ \t _ \t O \t _ \t\n _ \t P \t O \t O \t\n _ \t _ \t _ \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100005,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t G \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n P \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100006,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t G \t _ \t _ \t\n _ \t _ \t _ \t P \t\n _ \t _ \t _ \t _ \t\n _ \t O \t _ \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100007,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t P \t G \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100008,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t P \t\n G \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100009,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t _ \t O \t _ \t\n G \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n O \t _ \t P \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100010,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t P \t _ \t O \t\n O \t _ \t G \t O \t\n O \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100011,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t P \t _ \t\n G \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100012,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\n O \t _ \t _ \t _ \t\n P \t G \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100013,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t O \t _ \t\n G \t _ \t _ \t O \t\n _ \t O \t _ \t P \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100014,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t G \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t P \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100015,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n G \t O \t _ \t _ \t\n _ \t _ \t O \t O \t\n _ \t _ \t O \t _ \t\n O \t P \t _ \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100016,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t O \t _ \t\n P \t _ \t _ \t _ \t\n O \t G \t O \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100017,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t P \t _ \t O \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\n O \t O \t G \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100018,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t G \t _ \t O \t\n O \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n P \t O \t O \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100019,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t G \t\n _ \t _ \t P \t _ \t\n O \t _ \t O \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100020,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t O \t\n _ \t O \t G \t _ \t\n _ \t P \t _ \t _ \t\n O \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100021,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t O \t O \t\n _ \t _ \t _ \t P \t\n O \t G \t _ \t _ \t\n _ \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100022,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\n P \t _ \t G \t O \t\n _ \t _ \t _ \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100023,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t O \t\n _ \t _ \t G \t _ \t\n _ \t _ \t P \t _ \t\n O \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100024,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t O \t _ \t _ \t\n _ \t O \t G \t O \t\n P \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100025,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t P \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t O \t O \t\n _ \t _ \t G \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100026,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n O \t _ \t _ \t G \t\n _ \t _ \t O \t O \t\n _ \t _ \t P \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100027,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t P \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t O \t O \t _ \t\n _ \t _ \t G \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100028,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t O \t P \t _ \t\n _ \t _ \t _ \t _ \t\n G \t _ \t _ \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100029,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t O \t G \t\n _ \t _ \t _ \t _ \t\n O \t _ \t _ \t O \t\n _ \t P \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100030,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t P \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t G \t _ \t\n _ \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100031,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n G \t _ \t O \t P \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t O \t\n O \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100032,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n G \t _ \t P \t O \t\n _ \t _ \t _ \t _ \t\n O \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100033,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n P \t _ \t _ \t _ \t\n _ \t O \t _ \t _ \t\n O \t _ \t O \t _ \t\n _ \t _ \t _ \t G \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100034,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t P \t O \t\n _ \t _ \t _ \t _ \t\n O \t G \t _ \t _ \t\n O \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100035,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t G \t _ \t _ \t\n O \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t P \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100036,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t G \t P \t\n O \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100037,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t _ \t _ \t\n O \t _ \t _ \t _ \t\n _ \t _ \t G \t _ \t\n _ \t _ \t P \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100038,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t O \t\n _ \t P \t _ \t O \t\n _ \t _ \t O \t _ \t\n G \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100039,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t P \t _ \t O \t\n _ \t _ \t G \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100040,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t _ \t _ \t _ \t\n _ \t G \t _ \t _ \t\n O \t _ \t _ \t _ \t\n O \t P \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100041,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t O \t _ \t\n _ \t O \t _ \t _ \t\n P \t _ \t _ \t _ \t\n _ \t _ \t G \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100042,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\n G \t _ \t _ \t _ \t\n _ \t _ \t _ \t P \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100043,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t G \t _ \t\n O \t _ \t _ \t P \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100044,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t P \t _ \t O \t\n _ \t _ \t _ \t G \t\n _ \t O \t _ \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100045,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t O \t _ \t _ \t\n _ \t P \t G \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100046,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n G \t O \t O \t O \t\n _ \t P \t _ \t _ \t\n O \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100047,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t P \t O \t G \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100048,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t G \t O \t\n _ \t O \t O \t _ \t\n P \t _ \t O \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100049,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n P \t _ \t O \t _ \t\n _ \t _ \t O \t _ \t\n _ \t G \t O \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100050,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t O \t O \t\n _ \t _ \t G \t _ \t\n _ \t _ \t _ \t O \t\n _ \t _ \t _ \t P \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100051,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t G \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t O \t P \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100052,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t G \t\n O \t _ \t _ \t _ \t\n P \t _ \t O \t _ \t\n _ \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100053,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t O \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t O \t P \t\n _ \t G \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100054,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n P \t O \t _ \t O \t\n G \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\n O \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100055,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n G \t O \t _ \t _ \t\n _ \t _ \t _ \t P \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100056,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t O \t _ \t _ \t\n _ \t O \t P \t _ \t\n G \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100057,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t G \t\n _ \t _ \t O \t _ \t\n _ \t _ \t P \t _ \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100058,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t O \t _ \t G \t\n _ \t _ \t O \t _ \t\n _ \t _ \t _ \t _ \t\n O \t P \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100059,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t O \t O \t\n _ \t _ \t _ \t _ \t\n O \t G \t _ \t _ \t\n _ \t _ \t P \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100060,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t _ \t _ \t\n G \t _ \t O \t _ \t\n P \t _ \t _ \t O \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100061,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t O \t _ \t _ \t\n G \t _ \t O \t _ \t\n P \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100062,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t O \t _ \t _ \t\n O \t O \t _ \t _ \t\n _ \t G \t _ \t _ \t\n P \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100063,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t G \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\n O \t P \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100064,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n G \t _ \t _ \t _ \t\n _ \t _ \t P \t _ \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100065,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t P \t _ \t _ \t\n _ \t _ \t G \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100066,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n O \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n P \t _ \t _ \t G \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100067,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t G \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t O \t\n _ \t _ \t P \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100068,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t _ \t _ \t _ \t\n _ \t _ \t _ \t O \t\n G \t O \t _ \t O \t\n _ \t O \t P \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100069,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t P \t _ \t\n O \t _ \t O \t _ \t\n _ \t G \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100070,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t P \t _ \t\n _ \t _ \t _ \t O \t\n G \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100071,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t G \t P \t _ \t\n _ \t _ \t _ \t _ \t\n O \t _ \t _ \t _ \t\n O \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100072,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t _ \t _ \t\n _ \t O \t O \t O \t\n _ \t O \t _ \t _ \t\n _ \t P \t _ \t G \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100073,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n O \t _ \t P \t _ \t\n O \t G \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100074,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t P \t G \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100075,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t O \t _ \t O \t\n G \t _ \t P \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100076,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t O \t _ \t\n _ \t _ \t O \t _ \t\n G \t P \t O \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100077,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t P \t O \t\n O \t O \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t O \t G \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100078,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n P \t _ \t O \t O \t\n O \t _ \t _ \t _ \t\n _ \t _ \t _ \t O \t\n G \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100079,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n P \t _ \t _ \t G \t\n O \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n O \t O \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100080,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t _ \t P \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t G \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100081,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t P \t _ \t\n _ \t G \t _ \t _ \t\n _ \t _ \t _ \t O \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100082,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t P \t G \t _ \t\n _ \t O \t _ \t _ \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100083,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t P \t G \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t O \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100084,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t _ \t _ \t\n _ \t G \t _ \t P \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100085,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t G \t _ \t\n _ \t _ \t _ \t O \t\n _ \t _ \t _ \t _ \t\n P \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100086,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t _ \t _ \t\n _ \t _ \t O \t _ \t\n P \t _ \t _ \t G \t\n _ \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100087,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n G \t _ \t _ \t _ \t\n _ \t O \t _ \t _ \t\n O \t _ \t _ \t P \t\n _ \t _ \t _ \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100088,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t O \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n P \t G \t _ \t O \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100089,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t P \t _ \t\n _ \t G \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100090,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t _ \t O \t _ \t\n _ \t _ \t P \t _ \t\n _ \t O \t O \t _ \t\n _ \t _ \t G \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100091,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t P \t\n _ \t G \t _ \t O \t\n _ \t _ \t _ \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100092,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t P \t _ \t _ \t\n _ \t _ \t _ \t _ \t\n _ \t G \t O \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100093,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t _ \t _ \t _ \t\n _ \t _ \t P \t G \t\n _ \t _ \t _ \t _ \t\n O \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100094,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t G \t _ \t _ \t\n _ \t _ \t O \t _ \t\n _ \t _ \t _ \t P \t\n _ \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100095,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t _ \t O \t _ \t\n _ \t P \t _ \t _ \t\n _ \t G \t _ \t _ \t\n _ \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100096,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t O \t _ \t G \t\n P \t _ \t _ \t O \t\n _ \t _ \t O \t _ \t\n _ \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100097,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n O \t P \t _ \t O \t\n O \t _ \t _ \t _ \t\n G \t _ \t _ \t _ \t\n _ \t O \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100098,
"split": "train"
}
|
frozenlake
|
[
{
"content": "<|im_start|>user\nYou are walking on a frozen lake.\n\nFrozenLake Quick Guide\nGoal: Reach the goal (G).\n\nSymbols:\n_ Frozen | O Hole | G Goal | P Player\n\nRules:\n1. Avoid falling into holes (O).\n2. Frozen tiles are slippery, you may move perpendicular to your intended direction.\n\nAnswers:\n<answer> Up </answer> | <answer> Down </answer> | <answer> Left </answer> | <answer> Right </answer>\n\nRewards:\nFall into hole: 0\nReach goal: +1.0\n\n\n[Cumulative Observations]:\n _ \t P \t O \t O \t\n _ \t _ \t _ \t _ \t\n O \t _ \t _ \t _ \t\n G \t _ \t _ \t _ \t\nDecide the next action:\nAlways output: <think> [Your thoughts] </think> <answer> [your answer] </answer> with no extra text. Strictly follow this format. <|im_end|>\n<|im_start|>assistant\n<think>",
"role": "user"
}
] |
bfs
|
{
"ground_truth": {
"numbers": [
0,
0
],
"target": 0
},
"style": "rule"
}
|
{
"index": 100099,
"split": "train"
}
|
End of preview. Expand
in Data Studio
No dataset card yet
- Downloads last month
- 775