albert_chinese_xlarge/bert4torch_config.json

{
  "attention_probs_dropout_prob": 0,
  "bos_token_id": 2,
  "classifier_dropout_prob": 0.1,
  "down_scale_factor": 1,
  "embedding_size": 128,
  "eos_token_id": 3,
  "gap_size": 0,
  "hidden_act": "relu",
  "hidden_dropout_prob": 0,
  "hidden_size": 2048,
  "initializer_range": 0.014,
  "inner_group_num": 1,
  "intermediate_size": 8192,
  "layer_norm_eps": 1e-12,
  "layers_to_keep": [],
  "max_position_embeddings": 512,
  "model": "albert",
  "net_structure_type": 0,
  "num_attention_heads": 16,
  "num_hidden_groups": 1,
  "num_hidden_layers": 24,
  "num_memory_blocks": 0,
  "pad_token_id": 0,
  "type_vocab_size": 2,
  "vocab_size": 21128
}