config.json

{
  "architectures": [
    "ColQwenStella"
  ],
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoModel": "modeling_colqwenstella.ColQwenStella",
    "AutoConfig": "modeling_colqwenstella.ColStellaVLConfig"
  },
  "img_end_token": "<|im_end|>",
  "img_end_token_id": 151645,
  "img_start_token": "<|im_start|>",
  "img_start_token_id": 151644,
  "img_token": "<|image_pad|>",
  "img_token_id": 151646,
  "model_type": "colstella_vl",
  "text_config": {
    "_attn_implementation_autoset": true,
    "bos_token_id": 151643,
    "eos_token_id": 151643,
    "hidden_size": 1536,
    "intermediate_size": 8960,
    "max_position_embeddings": 131072,
    "max_window_layers": 21,
    "model_type": "qwen2",
    "num_attention_heads": 12,
    "num_hidden_layers": 28,
    "num_key_value_heads": 2,
    "rope_theta": 1000000.0,
    "torch_dtype": "float32",
    "vocab_size": 151647
  },
  "torch_dtype": "float32",
  "transformers_version": "4.46.3",
  "vision_config": {
    "_attn_implementation_autoset": true,
    "bos_token_id": 151643,
    "eos_token_id": 151645,
    "hidden_size": 1536,
    "image_token_id": 151646,
    "intermediate_size": 8960,
    "max_window_layers": 28,
    "model_type": "qwen2_vl",
    "num_attention_heads": 12,
    "num_hidden_layers": 28,
    "num_key_value_heads": 2,
    "rms_norm_eps": 1e-06,
    "rope_scaling": {
      "mrope_section": [
        16,
        24,
        24
      ],
      "rope_type": "default",
      "type": "default"
    },
    "sliding_window": 32768,
    "tie_word_embeddings": true,
    "torch_dtype": "float32",
    "video_token_id": 151656,
    "vision_config": {
      "hidden_size": 1536,
      "in_chans": 3,
      "spatial_patch_size": 14,
      "attn_implementation": "flash_attention_2"
    },
    "vision_end_token_id": 151653,
    "vision_start_token_id": 151652,
    "vision_token_id": 151654,
    "vocab_size": 151936
  }
}