{
  "_name_or_path": "./work_dirs/final_run/apollo-Qwen2.5-3B-Instruct-internvideo2-siglip-so400m-patch14-384-freeze-perciver_128_2-newprompt-ft/llm",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "max_position_embeddings": 32768,
  "max_window_layers": 70,
  "model_max_length": 16384,
  "model_type": "qwen2",
  "num_attention_heads": 16,
  "num_hidden_layers": 36,
  "num_key_value_heads": 2,
  "rms_norm_eps": 1e-06,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": true,
  "tokenizer_model_max_length": 16384,
  "tokenizer_padding_side": "right",
  "torch_dtype": "bfloat16",
  "transformers_version": "4.44.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151936
}