{
  "_name_or_path": "tiiuae/falcon-11B-vlm",
  "architectures": [
    "LlavaNextForConditionalGeneration"
  ],
  "hidden_size": 4096,
  "ignore_index": -100,
  "image_grid_pinpoints": [
    [
      336,
      672
    ],
    [
      672,
      336
    ],
    [
      672,
      672
    ],
    [
      1008,
      336
    ],
    [
      336,
      1008
    ]
  ],
  "image_token_index": 65024,
  "model_type": "llava_next",
  "pad_token_id": 0,
  "projector_hidden_act": "gelu",
  "text_config": {
    "_name_or_path": "tiiuae/falcon-11B-vlm",
    "architectures": [
      "FalconForCausalLM"
    ],
    "auto_map": {
      "AutoConfig": "configuration_falcon.FalconConfig",
      "AutoModel": "modeling_falcon.FalconModel",
      "AutoModelForCausalLM": "modeling_falcon.FalconForCausalLM",
      "AutoModelForQuestionAnswering": "modeling_falcon.FalconForQuestionAnswering",
      "AutoModelForSequenceClassification": "modeling_falcon.FalconForSequenceClassification",
      "AutoModelForTokenClassification": "modeling_falcon.FalconForTokenClassification"
    },
    "bias": false,
    "bos_token_id": 11,
    "eos_token_id": 11,
    "ff_factor": 4,
    "ffn_hidden_size": 16384,
    "hidden_dropout": 0.0,
    "hidden_size": 4096,
    "initializer_range": 0.02,
    "layer_norm_epsilon": 1e-05,
    "max_position_embeddings": 8192,
    "max_length": 4096,
    "model_type": "falcon",
    "multi_query": true,
    "new_decoder_architecture": true,
    "num_attention_heads": 32,
    "num_hidden_layers": 60,
    "num_kv_heads": 8,
    "num_ln_in_parallel_attn": 1,
    "parallel_attn": true,
    "rope_scaling": null,
    "rope_theta": 500042.0,
    "tie_word_embeddings": false,
    "torch_dtype": "bfloat16",
    "transformers_version": "4.39.2",
    "use_cache": true,
    "vocab_size": 65024
  },
  "torch_dtype": "bfloat16",
  "transformers_version": "4.41.0.dev0",
  "use_image_newline_parameter": true,
  "vision_config": {
    "hidden_size": 1024,
    "image_size": 336,
    "intermediate_size": 4096,
    "model_type": "clip_vision_model",
    "num_attention_heads": 16,
    "num_hidden_layers": 24,
    "patch_size": 14,
    "projection_dim": 768,
    "vocab_size": 32000
  },
  "vision_feature_layer": -2,
  "vision_feature_select_strategy": "default",
  "vocab_size": 65024
}