{
  "architectures": [
    "PolyverseForConditionalGeneration"
  ],
  "hidden_size": 1152,
  "image_token_index": 256204,
  "model_type": "Polyverse",
  "projection_dim": 1024,
  "sampler_tokens": 128,
  "text_config": {
    "_name_or_path": "facebook/nllb-200-distilled-1.3B",
    "architectures": [
      "M2M100ForConditionalGeneration"
    ],
    "decoder_ffn_dim": 8192,
    "decoder_layerdrop": 0,
    "decoder_layers": 24,
    "encoder_ffn_dim": 8192,
    "encoder_layerdrop": 0,
    "encoder_layers": 24,
    "max_length": 200,
    "model_type": "m2m_100",
    "num_hidden_layers": 24,
    "num_image_tokens": 729,
    "torch_dtype": "float32",
    "vocab_size": 256206
  },
  "torch_dtype": "float32",
  "transformers_version": "4.44.0",
  "vision_config": {
    "_name_or_path": "google/siglip-so400m-patch14-384",
    "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,
    "projection_dim": 1024
  }
}