{
  "architectures": [
    "SarmaeModel"
  ],
  "attention_probs_dropout_prob": 0.0,
  "checkpoint_stage": "pretrain",
  "dtype": "float32",
  "global_pool": true,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 768,
  "id2label": {},
  "image_mean": [
    0.485,
    0.456,
    0.406
  ],
  "image_size": 224,
  "image_std": [
    0.229,
    0.224,
    0.225
  ],
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "label2id": {},
  "layer_norm_eps": 1e-06,
  "mlp_ratio": 4.0,
  "model_type": "sarmae",
  "num_attention_heads": 12,
  "num_channels": 3,
  "num_hidden_layers": 12,
  "patch_size": 16,
  "qkv_bias": true,
  "repeat_grayscale_channels": true,
  "transformers_version": "5.0.0",
  "auto_map": {
    "AutoConfig": "modeling_sarmae.SarmaeConfig",
    "AutoModel": "modeling_sarmae.SarmaeModel",
    "AutoModelForImageClassification": "modeling_sarmae.SarmaeForImageClassification"
  },
  "custom_pipelines": {
    "sarmae-feature-extraction": {
      "impl": "pipeline_sarmae.SarmaeImageFeatureExtractionPipeline",
      "pt": [
        "AutoModel"
      ]
    }
  }
}