config.json

{
  "architectures": [
    "LlavaNextVideoForConditionalGeneration"
  ],
  "ignore_index": -100,
  "image_grid_pinpoints": [
    [
      336,
      672
    ],
    [
      672,
      336
    ],
    [
      672,
      672
    ],
    [
      1008,
      336
    ],
    [
      336,
      1008
    ]
  ],
  "image_token_index": 32001,
  "model_type": "llava-next-video",
  "projector_hidden_act": "gelu",
  "spatial_pool_mode": "average",
  "spatial_pool_stride": 2,
  "text_config": {
    "_name_or_path": "lmsys/vicuna-7b-v1.5",
    "architectures": [
      "LlamaForCausalLM"
    ],
    "factor": 2.0,
    "max_position_embeddings": 4096,
    "model_type": "llama",
    "pad_token_id": 0,
    "rms_norm_eps": 1e-05,
    "torch_dtype": "float16",
    "type": "linear",
    "vocab_size": 32064
  },
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.42.0.dev0",
  "use_image_newline_parameter": true,
  "video_token_index": 32000,
  "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"
}