config.json

{
  "architectures": [
    "LongLlamaForCausalLM"
  ],
  "bos_token_id": 1,
  "eos_token_id": 2,
  "gradient_checkpoint_every_ith": 1,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "last_context_length": 2048,
  "max_position_embeddings": 4096,
  "mem_attention_grouping": null,
  "mem_dtype": "bfloat16",
  "mem_layers": [
    8,
    16,
    24
  ],
  "mem_positionals": true,
  "model_type": "longllama",
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "pad_token_id": 0,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "tie_word_embeddings": false,
  "torch_attention": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.30.0",
  "use_cache": true,
  "vocab_size": 32016,
  "auto_map": {
    "AutoConfig": "configuration_longllama.LongLlamaConfig",
    "AutoModel": "modeling_longllama.LongLlamaModel",
    "AutoModelForCausalLM": "modeling_longllama.LongLlamaForCausalLM",
    "AutoModelForSequenceClassification": "modeling_longllama.LongLlamaForSequenceClassification"
  }
}