config.json

{
  "_name_or_path": "Xenova/tiny-random-LlavaForConditionalGeneration",
  "architectures": [
    "LlavaForConditionalGeneration"
  ],
  "ignore_index": -100,
  "image_seq_length": 225,
  "image_token_index": 32000,
  "model_type": "llava",
  "projector_hidden_act": "gelu",
  "text_config": {
    "_name_or_path": "HuggingFaceM4/tiny-random-LlamaForCausalLM",
    "architectures": [
      "LlamaForCausalLM"
    ],
    "bos_token_id": 0,
    "eos_token_id": 1,
    "hidden_size": 16,
    "intermediate_size": 64,
    "model_type": "llama",
    "num_attention_heads": 4,
    "num_hidden_layers": 2,
    "num_key_value_heads": 4,
    "pad_token_id": 2,
    "torch_dtype": "float32",
    "vocab_size": 32002
  },
  "torch_dtype": "float32",
  "transformers_version": "4.39.3",
  "vision_config": {
    "attention_dropout": 0.1,
    "dropout": 0.1,
    "hidden_size": 32,
    "image_size": 30,
    "intermediate_size": 37,
    "model_type": "clip_vision_model",
    "num_attention_heads": 4,
    "num_hidden_layers": 2,
    "patch_size": 2,
    "projection_dim": 32
  },
  "vision_feature_layer": -2,
  "vision_feature_select_strategy": "default"
}