{
    "_name_or_path": "./",
    "architectures": [
        "InfiMMVicunaModel"
    ],
    "auto_map": {
        "AutoConfig": "configuration_infimm_vicuna.InfiMMConfig",
        "AutoModelForCausalLM": "modeling_infimm_vicuna.InfiMMVicunaModel"
    },
    "model_type": "infimm-vicuna",
    "seq_length": 1024,
    "tokenizer_type": "LlamaTokenizer",
    "torch_dtype": "bfloat16",
    "transformers_version": "4.34.0",
    "use_cache": true,
    "use_flash_attn": false,
    "cross_attn_every_n_layers": 4,
    "use_grad_checkpoint": false,
    "freeze_llm": true,
    "image_token_id": 32001,
    "eoc_token_id": 32000,
    "visual": {
        "image_size": 224,
        "layers": 64,
        "width": 1792,
        "head_width": 112,
        "mlp_ratio": 8.571428571428571,
        "patch_size": 14,
        "eva_model_name": "eva-clip-4b-14-x",
        "drop_path_rate": 0,
        "xattn": false,
        "postnorm": true,
        "fusedLN": false,
        "embed_dim": 1024,
        "patch_dropout": 0
    },
    "language": {
        "_name_or_path": "lmsys/vicuna-13b-v1.5",
        "architectures": [
          "LlamaForCausalLM"
        ],
        "bos_token_id": 1,
        "eos_token_id": 2,
        "hidden_act": "silu",
        "hidden_size": 5120,
        "initializer_range": 0.02,
        "intermediate_size": 13824,
        "max_length": 4096,
        "max_position_embeddings": 4096,
        "model_type": "llama",
        "num_attention_heads": 40,
        "num_hidden_layers": 40,
        "num_key_value_heads": 40,
        "pad_token_id": 0,
        "pretraining_tp": 1,
        "rms_norm_eps": 1e-05,
        "rope_scaling": null,
        "tie_word_embeddings": false,
        "torch_dtype": "float16",
        "transformers_version": "4.34.0",
        "use_cache": true,
        "vocab_size": 32002
      }
  }