{
  "architectures": [
    "Bert2DForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0.1,
  "gradient_checkpointing": false,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "layer_norm_eps": 1e-12,
  "max_position_embeddings": 512,
  "model_type": "bert2d",
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "pad_token_id": 0,
  "max_word_position_embeddings": 512,
  "max_intermediate_subword_position_embeddings": 2,
  "position_embedding_type": "absolute",
  "type_vocab_size": 2,
  "use_cache": false,
  "vocab_size": 128000,
  "auto_map": {
    "AutoConfig": "configuration_bert2d.Bert2DConfig",
    "AutoModel": "modeling_bert2d.Bert2DModel",
    "AutoModelForMaskedLM": "modeling_bert2d.Bert2DForMaskedLM",
    "AutoTokenizer": [
      "tokenization_bert2d.Bert2DTokenizer",
      "tokenization_bert2d_fast.Bert2DTokenizerFast"
    ]
  },
  "tokenizer_class": "tokenization_bert2d.Bert2DTokenizer",
  "tokenizer_class_fast": "tokenization_bert2d_fast.Bert2DTokenizerFast"
}