{
  "_name_or_path": "checkpoints/Video-Panda-7B/videopanda_prtr1",
  "architectures": ["VideoPandaLlamaForCausalLM"],
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "image_aspect_ratio": "square",
  "image_grid_pinpoints": null,
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "max_position_embeddings": 4096,
  "mm_hidden_size": 1024,
  "mm_projector_lr": 2e-05,
  "mm_projector_type": "mlp2x_gelu",
  "mm_use_im_patch_token": false,
  "mm_use_im_start_end": false,
  "mm_video_tower": "videopanda/config/videopanda-video-LB-patch14-anypixel-448",
  "mm_video_tower_teacher": "LanguageBind/LanguageBind_Video_merge",
  "mm_vision_select_feature": "patch",
  "mm_vision_select_layer": -2,
  "mm_vision_tower": "videopanda/config/videopanda-vision-LB-patch14-anypixel-448",
  "mm_vision_tower_teacher": "LanguageBind/LanguageBind_Image",
  "model_type": "videopanda",
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": 32,
  "pad_token_id": 0,
  "pretraining_tp": 1,
  "requires_video_distill": true,
  "requires_image_distill": false,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.31.0",
  "tune_mm_mlp_adapter": false,
  "use_cache": true,
  "use_mm_proj": true,
  "video_hidden_size": 1024,
  "video_projector_lr": 2e-05,
  "video_tower_lr": 2e-05,
  "video_tower_teacher": "LanguageBind/LanguageBind_Video_merge",
  "vision_tower_lr": 2e-05,
  "vocab_size": 32000
}