config.json

{
  "_name_or_path": "./llama_fuyu.json",
  "architectures": [
    "FuyuForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "relu2",
  "hidden_dropout": 0.0,
  "hidden_size": 2048,
  "image_size": 300,
  "initializer_range": 0.02,
  "intermediate_size": 16384,
  "layer_norm_eps": 1e-05,
  "max_position_embeddings": 16384,
  "model_type": "fuyu",
  "num_attention_heads": 64,
  "num_channels": 3,
  "num_hidden_layers": 36,
  "partial_rotary_factor": 0.5,
  "patch_size": 30,
  "qk_layernorm": true,
  "rope_scaling": null,
  "rope_theta": 25000.0,
  "text_config": {
    "architectures": [
      "LlamaForCausalLM"
    ],
    "attention_bias": false,
    "hidden_act": "silu",
    "hidden_size": 2048,
    "intermediate_size": 5632,
    "max_position_embeddings": 2048,
    "model_type": "llama",
    "num_attention_heads": 32,
    "num_hidden_layers": 22,
    "num_key_value_heads": 4,
    "pretraining_tp": 1,
    "rms_norm_eps": 1e-05,
    "rope_theta": 10000.0,
    "torch_dtype": "float32",
    "vocab_size": 32001
  },
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.36.2",
  "use_cache": true,
  "vocab_size": 32001
}