sijunhe/nezha-base-wwm/bert4torch_config.json

{
  "attention_probs_dropout_prob": 0.1,
  "bos_token_id": 2,
  "classifier_dropout": 0.1,
  "embedding_size": 128,
  "eos_token_id": 3,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "inner_group_num": 1,
  "intermediate_size": 3072,
  "layer_norm_eps": 1e-12,
  "max_position_embeddings": 512,
  "max_relative_position": 64,
  "model": "nezha",
  "num_attention_heads": 12,
  "num_hidden_groups": 1,
  "num_hidden_layers": 12,
  "pad_token_id": 0,
  "torch_dtype": "float32",
  "type_vocab_size": 2,
  "use_cache": true,
  "use_relative_position": true,
  "vocab_size": 21128
}