{
 "architectures": [
  "BertForMaskedLM"
 ],
 "model_type": "bert",
 "hidden_size": 32,
 "num_hidden_layers": 2,
 "num_attention_heads": 2,
 "intermediate_size": 64,
 "vocab_size": 100,
 "max_position_embeddings": 64,
 "type_vocab_size": 2,
 "pad_token_id": 0,
 "layer_norm_eps": 1e-12,
 "hidden_act": "gelu",
 "attention_probs_dropout_prob": 0.1,
 "hidden_dropout_prob": 0.1,
 "initializer_range": 0.02
}