config.json

{
  "architectures": [
    "GrinQwen2VLForConditionalGeneration"
  ],
  "attention_dropout": 0.0,
  "audio_end_token_id": 151666,
  "audio_start_token_id": 151665,
  "audio_token_id": 151667,
  "avg_hidden_states_last": false,
  "bos_token_id": 151643,
  "capacity_factor": 3.0,
  "drop_policy": "probs",
  "drop_token_num_print": true,
  "dynamic_intermediate_size": 18944,
  "enable_expert_tensor_parallelism": false,
  "eos_token_id": 151645,
  "ep_size": 1,
  "fixed_ep_size": 1,
  "fp32_gate": true,
  "hidden_act": "silu",
  "hidden_size": 3584,
  "ignore_differentiable_router": false,
  "image_grid_pinpoints": [
    [
      384,
      384
    ],
    [
      384,
      768
    ],
    [
      384,
      1152
    ],
    [
      384,
      1536
    ],
    [
      768,
      384
    ],
    [
      768,
      768
    ],
    [
      768,
      1152
    ],
    [
      768,
      1536
    ],
    [
      1152,
      384
    ],
    [
      1152,
      768
    ],
    [
      1152,
      1152
    ],
    [
      1152,
      1536
    ],
    [
      1536,
      384
    ],
    [
      1536,
      768
    ],
    [
      1536,
      1152
    ],
    [
      1536,
      1536
    ]
  ],
  "image_token_id": 151655,
  "initializer_range": 0.02,
  "input_jitter_noise": 0.01,
  "intermediate_size": 18944,
  "l_aux_weight": 0.025,
  "l_aux_weight_decay_steps": 100000,
  "layer_types": [
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention"
  ],
  "max_position_embeddings": 128000,
  "max_window_layers": 28,
  "min_capacity": 6,
  "min_l_aux_weight": 0.025,
  "mlp_dynamic_expert_num": 4,
  "mlp_dynamic_null_expert_num": 1,
  "mlp_dynamic_top_k": 2,
  "mlp_dynamic_top_p": 0.7,
  "mlp_fixed_expert_num": 2,
  "mm_hidden_size": 1152,
  "model_type": "grin_qwen2_vl",
  "num_attention_heads": 28,
  "num_hidden_layers": 28,
  "num_key_value_heads": 4,
  "rms_norm_eps": 1e-06,
  "rope_scaling": {
    "mrope_section": [
      16,
      24,
      24
    ],
    "rope_type": "default",
    "type": "default"
  },
  "rope_theta": 1000000.0,
  "router_jitter_noise": 0.01,
  "shared_intermediate_size": 2368,
  "sliding_window": 32768,
  "speech_end_token_id": 151669,
  "speech_prompt_token_id": 151670,
  "speech_split_token_id": 151671,
  "speech_start_token_id": 151668,
  "tie_word_embeddings": false,
  "token_drop": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.45.1",
  "tune_image_generator": false,
  "tune_speech_generator": false,
  "tune_speech_generator_only": false,
  "use_cache": true,
  "use_sliding_window": false,
  "video_token_id": 151656,
  "vision_config": {
    "hidden_size": 1152,
    "image_size": 384,
    "intermediate_size": 4304,
    "model_type": "siglip_vision_model",
    "num_attention_heads": 16,
    "num_hidden_layers": 27,
    "patch_size": 14
  },
  "audio_config": {
    "activation_dropout": 0.0,
    "activation_function": "gelu",
    "apply_spec_augment": false,
    "architectures": [
      "WhisperForConditionalGeneration"
    ],
    "attention_dropout": 0.0,
    "begin_suppress_tokens": [
      220,
      50257
    ],
    "bos_token_id": 50257,
    "classifier_proj_size": 256,
    "d_model": 1280,
    "decoder_attention_heads": 20,
    "decoder_ffn_dim": 5120,
    "decoder_layerdrop": 0.0,
    "decoder_layers": 32,
    "decoder_start_token_id": 50258,
    "dropout": 0.0,
    "encoder_attention_heads": 20,
    "encoder_ffn_dim": 5120,
    "encoder_layerdrop": 0.0,
    "encoder_layers": 32,
    "eos_token_id": 50257,
    "init_std": 0.02,
    "is_encoder_decoder": true,
    "mask_feature_length": 10,
    "mask_feature_min_masks": 0,
    "mask_feature_prob": 0.0,
    "mask_time_length": 10,
    "mask_time_min_masks": 2,
    "mask_time_prob": 0.05,
    "max_length": 448,
    "max_source_positions": 1500,
    "max_target_positions": 448,
    "median_filter_width": 7,
    "model_type": "whisper",
    "num_hidden_layers": 32,
    "num_mel_bins": 128,
    "pad_token_id": 50256,
    "scale_embedding": false,
    "torch_dtype": "float16",
    "transformers_version": "4.36.0.dev0",
    "use_cache": true,
    "use_weighted_layer_sum": false,
    "vocab_size": 51866
  },
  "vision_end_token_id": 151653,
  "vision_hidden_size": 1152,
  "vision_spatial_pool_stride": 12,
  "vision_start_token_id": 151652,
  "vision_token_id": 151654,
  "vision_tower_pretrained": null,
  "vision_vision_select_layer": -2,
  "visual_aligner_type": "linear",
  "vocab_size": 152064,
  "whisper_hidden_size": 1280,
  "whisper_projector_type": "linear",
  "whisper_query_tokens_size": 200
}