{
  "_name_or_path": "Griffon-G-Llama2-7B",
  "architectures": [
    "GriffonLlama2ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "freeze_llm": false,
  "freeze_mm_mlp_adapter": false,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "image_aspect_ratio": "square",
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "max_position_embeddings": 4096,
  "mlp_bias": false,
  "mm_hidden_size": 1024,
  "mm_projector_lr": null,
  "mm_projector_type": "conv_reduce",
  "mm_use_im_patch_token": false,
  "mm_use_im_start_end": false,
  "mm_vision_select_feature": "patch",
  "mm_vision_select_layer": -2,
  "mm_vision_tower": "JefferyZhan/clip-vit-large-path14-336_to_1022",
  "model_type": "llama2_Griffon",
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": 32,
  "pad_token_id": 0,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "single_num": false,
  "tie_word_embeddings": false,
  "tokenizer_model_max_length": 3072,
  "tokenizer_padding_side": "right",
  "torch_dtype": "bfloat16",
  "transformers_version": "4.50.0.dev0",
  "tune_mm_mlp_adapter": false,
  "unshared_visual_encoder": false,
  "use_cache": null,
  "use_mm_proj": true,
  "vocab_size": 32000
}