vision_tower/internvideo2/config.json

{
  "architectures": [
    "PretrainInternVideo2"
  ],
  "attention_probs_dropout_prob": 0.0,
  "attn_pool_num_heads": 16,
  "checkpoint_num": 40,
  "clip_embed_dim": 768,
  "clip_input_resolution": 224,
  "clip_norm_type": "l2",
  "clip_return_layer": 6,
  "clip_student_return_interval": 1,
  "clip_teacher": null,
  "clip_teacher_embed_dim": 3200,
  "clip_teacher_final_dim": 768,
  "clip_teacher_return_interval": 1,
  "d_model": 1408,
  "encoder_stride": 16,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 768,
  "image_mask_ratio": 0.5,
  "image_mask_type": "random",
  "image_size": 224,
  "img_size": 224,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "keep_temporal": false,
  "layer_norm_eps": 1e-12,
  "model_type": "internvideo2",
  "name": "pretrain_internvideo2_1b_patch14_224",
  "num_attention_heads": 12,
  "num_channels": 3,
  "num_frames": 4,
  "num_heads": 16,
  "num_hidden_layers": 12,
  "only_mask": true,
  "patch_size": 14,
  "qkv_bias": false,
  "sep_image_video_pos_embed": true,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.44.0",
  "tubelet_size": 1,
  "use_checkpoint": true,
  "use_flash_attn": false,
  "use_fused_mlp": false,
  "use_fused_rmsnorm": false,
  "video_mask_ratio": 0.8,
  "video_mask_type": "random",
  "auto_map": {
    "AutoConfig": "configuration_internvideo2.InternVideo2Config",
    "AutoModel": "modeling_internvideo2.InternVideo2Model"
  }
}