{
    "_attn_implementation_autoset": true,
    "_name_or_path": "PKU-Alignment/Align-DS-V",
    "architectures": [
      "LlavaForConditionalGeneration"
    ],
    "bos_token_id": 128000,
    "eos_token_id": 128001,
    "ignore_index": -100,
    "image_seq_length": 576,
    "image_token_index": 128256,
    "model_type": "llava",
    "multimodal_projector_bias": true,
    "pad_token_id": 128258,
    "projector_hidden_act": "gelu",
    "text_config": {
      "_attn_implementation_autoset": true,
      "_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "architectures": [
        "LlamaForCausalLM"
      ],
      "bos_token_id": 128000,
      "eos_token_id": 128001,
      "intermediate_size": 14336,
      "max_position_embeddings": 131072,
      "model_type": "llama",
      "num_key_value_heads": 8,
      "rms_norm_eps": 1e-05,
      "rope_scaling": {
        "factor": 8.0,
        "high_freq_factor": 4.0,
        "low_freq_factor": 1.0,
        "original_max_position_embeddings": 8192,
        "rope_type": "llama3"
      },
      "rope_theta": 500000.0,
      "torch_dtype": "bfloat16",
      "vocab_size": 128259
    },
    "torch_dtype": "bfloat16",
    "transformers_version": "4.48.1",
    "vision_config": {
      "_attn_implementation_autoset": true,
      "dropout": 0.0,
      "hidden_size": 1024,
      "image_size": 336,
      "intermediate_size": 4096,
      "model_type": "clip_vision_model",
      "num_attention_heads": 16,
      "num_hidden_layers": 24,
      "patch_size": 14,
      "projection_dim": 768
    },
    "vision_feature_layer": -2,
    "vision_feature_select_strategy": "default"
  }