{
  "architectures": [
    "AlbertForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0,
  "hidden_act": "gelu_new",
  "hidden_dropout_prob": 0,
  "embedding_size": 128,
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 16384,
  "max_position_embeddings": 512,
  "num_attention_heads": 64,
  "num_hidden_layers": 12,
  "num_hidden_groups": 1,
  "net_structure_type": 0,
  "layers_to_keep": [],
  "gap_size": 0,
  "num_memory_blocks": 0,
  "inner_group_num": 1,
  "down_scale_factor": 1,
  "type_vocab_size": 2,
  "vocab_size": 30000
}