config.json

{
  "architectures": [
    "InternVLForConditionalGeneration"
  ],
  "downsample_ratio": 0.5,
  "image_seq_length": 256,
  "image_token_id": 151671,
  "model_type": "internvl",
  "projector_hidden_act": "gelu",
  "text_config": {
    "_name_or_path": "/root/codespace/checkpoints/Qwen3-4B",
    "architectures": [
      "Qwen3ForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 151643,
    "debug": false,
    "eos_token_id": 151645,
    "ep_size": 1,
    "head_dim": 128,
    "hidden_act": "silu",
    "hidden_size": 2560,
    "initializer_range": 0.02,
    "intermediate_size": 9728,
    "layer_types": [
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention"
    ],
    "max_position_embeddings": 40960,
    "max_window_layers": 36,
    "micro_forward": false,
    "model_type": "qwen3",
    "num_attention_heads": 32,
    "num_hidden_layers": 36,
    "num_key_value_heads": 8,
    "rms_norm_eps": 1e-06,
    "rope_scaling": null,
    "rope_theta": 1000000,
    "skip_checkpoint": false,
    "sliding_window": null,
    "torch_dtype": "bfloat16",
    "use_cache": true,
    "use_deepep": false,
    "use_sliding_window": false,
    "vocab_size": 151936
  },
  "torch_dtype": "bfloat16",
  "transformers_version": "4.55.0",
  "vision_config": {
    "architectures": [
      "InternVisionModel"
    ],
    "attention_bias": true,
    "attention_dropout": 0.0,
    "dropout": 0.0,
    "hidden_act": "gelu",
    "hidden_dropout_prob": 0.0,
    "hidden_size": 1024,
    "image_size": [
      448,
      448
    ],
    "initializer_factor": 0.1,
    "initializer_range": 1e-10,
    "intermediate_size": 4096,
    "layer_norm_eps": 1e-06,
    "layer_scale_init_value": 0.1,
    "model_type": "internvl_vision",
    "norm_type": "layer_norm",
    "num_attention_heads": 16,
    "num_channels": 3,
    "num_hidden_layers": 24,
    "patch_size": [
      14,
      14
    ],
    "projection_dropout": 0.0,
    "torch_dtype": "bfloat16",
    "use_absolute_position_embeddings": true,
    "use_mask_token": false,
    "use_mean_pooling": true,
    "use_qk_norm": false
  },
  "vision_feature_layer": -1,
  "vision_feature_select_strategy": "default"
}