ppllava_vicuna7b_image_video/config.json

{
  "architectures": [
    "LlavaInterleaveForConditionalGeneration"
  ],
  "btadapter": true,
  "btadapter_depth": 4,
  "clip_post_pretrain": null,
  "clip_weight": "openai/clip-vit-large-patch14-336",
  "frame_shape": [
    24,
    24
  ],
  "hidden_size": 4096,
  "ignore_index": -100,
  "image_grid_pinpoints": [
    [
      336,
      672
    ],
    [
      672,
      336
    ],
    [
      672,
      672
    ],
    [
      1008,
      336
    ],
    [
      336,
      1008
    ]
  ],
  "image_pooling_kernel": [
    1,
    3,
    3
  ],
  "image_pooling_stride": [
    1,
    3,
    3
  ],
  "image_token_index": 32000,
  "long_clip": true,
  "max_T": 64,
  "model_type": "llava_next",
  "pad_token_id": 0,
  "pllava_pooling_shape": null,
  "pooling": "ppllava",
  "pooling_kernel": [
    2,
    3,
    3
  ],
  "pooling_stride": [
    2,
    3,
    3
  ],
  "pooling_temp": 0.01,
  "projector_hidden_act": "gelu",
  "text_config": {
    "_name_or_path": "lmsys/vicuna-7b-v1.5",
    "architectures": [
      "LlamaForCausalLM"
    ],
    "max_position_embeddings": 4096,
    "model_type": "llama",
    "pad_token_id": 0,
    "rms_norm_eps": 1e-05,
    "torch_dtype": "float16",
    "vocab_size": 32064
  },
  "torch_dtype": "bfloat16",
  "transformers_version": "4.40.0",
  "use_image_newline_parameter": true,
  "video_input": "mean",
  "vision_config": {
    "depth": 4,
    "hidden_size": 1024,
    "image_size": 336,
    "intermediate_size": 4096,
    "max_T": 64,
    "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": 32064
}