{
  "_name_or_path": "/scratch/hpc162a02/cvpr/huggingface/hub/LongMamba_V4/LongMamba-clip-Stage12-2.7b/",
  "add_faster_video": false,
  "add_time_instruction": true,
  "architectures": [
    "LlavaMambaForCausalLM"
  ],
  "attn_cfg": {},
  "attn_layer_idx": [],
  "d_intermediate": 0,
  "d_model": 2560,
  "faster_token_stride": 10,
  "force_sample": false,
  "fused_add_norm": true,
  "image_aspect_ratio": "square",
  "image_crop_resolution": null,
  "image_grid_pinpoints": null,
  "image_split_resolution": null,
  "label_top_k": 5,
  "mm_hidden_size": 1024,
  "mm_newline_position": "grid",
  "mm_patch_merge_type": "flat",
  "mm_projector_lr": null,
  "mm_projector_type": "mlp2x_gelu",
  "mm_spatial_pool_mode": "bilinear",
  "mm_spatial_pool_stride": null,
  "mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_language_model",
  "mm_use_im_patch_token": false,
  "mm_use_im_start_end": false,
  "mm_vision_select_feature": "patch",
  "mm_vision_select_layer": -2,
  "mm_vision_tower": "openai/clip-vit-large-patch14",
  "mm_vision_tower_lr": 4e-06,
  "model_type": "llava_mamba",
  "n_layer": 64,
  "pad_vocab_size_multiple": 16,
  "pos_skipping_range": 4096,
  "residual_in_fp32": true,
  "rms_norm": true,
  "ssm_cfg": {
    "layer": "Mamba2"
  },
  "tie_embeddings": true,
  "tokenizer_model_max_length": 1048576,
  "tokenizer_padding_side": "right",
  "torch_dtype": "bfloat16",
  "transformers_version": "4.44.2",
  "use_cache": true,
  "use_mm_proj": true,
  "use_pos_skipping": false,
  "vision_tower_pretrained": null,
  "vocab_size": 50277
}