config.json

{
 "architectures": [
  "JasperVL"
 ],
 "auto_map": {
  "AutoModel": "modeling_jasper_vl.JasperVL",
  "AutoConfig": "configuration_jasper_vl.JasperVLConfig"
 },
 "img_end_token": "<|jasper_img_end|>",
 "img_end_token_id": 151648,
 "img_start_token": "<|jasper_img_start|>",
 "img_start_token_id": 151646,
 "img_token": "<|jasper_img_token|>",
 "img_token_id": 151647,
 "is_text_encoder": false,
 "model_type": "jasper_vl",
 "num_img_tokens": 300,
 "text_config": {
  "_attn_implementation_autoset": true,
  "architectures": [
   "JasperVL"
  ],
  "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": 151649
 },
 "torch_dtype": "bfloat16",
 "transformers_version": "4.46.1",
 "vector_dim": 12288,
 "vector_dropout_p": 0.2,
 "vision_config": {
  "_attn_implementation_autoset": true,
  "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
 }
}