Baseline + Manager/config.json

{
  "_name_or_path": "/share/home/xuxiao/model/checkpoint/LLaVA-Manager/llavanext-google_siglip-so400m-patch14-384-Qwen_Qwen2-0.5B-Instruct-mlp2x_gelu-stage1.5-manager-nogrid-1e-3-5e-5-static_zerouni-all-13-1-26-0-4-24-True",
  "add_faster_video": false,
  "add_time_instruction": false,
  "architectures": [
    "LlavaQwenManagerForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "faster_token_stride": 10,
  "force_sample": false,
  "hidden_act": "silu",
  "hidden_size": 896,
  "image_aspect_ratio": "square",
  "image_crop_resolution": null,
  "image_grid_pinpoints": null,
  "image_split_resolution": null,
  "initializer_range": 0.02,
  "intermediate_size": 4864,
  "max_position_embeddings": 32768,
  "max_window_layers": 24,
  "mm_hidden_size": 1152,
  "mm_manager_grid_type": "all",
  "mm_manager_index_list": [
    0,
    4,
    8,
    12,
    16,
    20
  ],
  "mm_manager_injection_end": 24,
  "mm_manager_injection_interval": 4,
  "mm_manager_injection_start": 0,
  "mm_manager_lr": 5e-05,
  "mm_manager_residual": true,
  "mm_manager_select_layer_index_list": [
    13,
    14,
    15,
    16,
    17,
    18,
    19,
    20,
    21,
    22,
    23,
    24,
    25
  ],
  "mm_manager_softmax_temperature": 1.0,
  "mm_manager_type": "static_zerouni",
  "mm_manager_vision_select_layers_end": 26,
  "mm_manager_vision_select_layers_interval": 1,
  "mm_manager_vision_select_layers_start": 13,
  "mm_newline_position": "grid",
  "mm_patch_merge_type": "flat",
  "mm_projector_lr": null,
  "mm_projector_type": "mlp2x_gelu",
  "mm_resampler_type": null,
  "mm_spatial_pool_mode": "bilinear",
  "mm_spatial_pool_stride": null,
  "mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_manager,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": "/share/home/xuxiao/model/siglip-so400m-patch14-384",
  "mm_vision_tower_lr": 2e-06,
  "model_type": "qwen2",
  "num_attention_heads": 14,
  "num_hidden_layers": 24,
  "num_key_value_heads": 2,
  "pos_skipping_range": 4096,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": true,
  "tokenizer_model_max_length": 16384,
  "tokenizer_padding_side": "right",
  "torch_dtype": "bfloat16",
  "transformers_version": "4.40.0.dev0",
  "use_cache": true,
  "use_mm_proj": true,
  "use_pos_skipping": false,
  "use_sliding_window": false,
  "vision_tower_pretrained": null,
  "vocab_size": 151936
}