bert4torch_config / mt5-base /bert4torch_config.json
Tongjilibo's picture
添加配置文件
36f7606
{
"model": "mt5.1.1",
"segment_vocab_size": 0,
"attention_probs_dropout_prob": 0.1,
"hidden_act": "gelu_new",
"hidden_dropout_prob": 0.1,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 2048,
"max_position_embeddings": 512,
"num_attention_heads": 12,
"num_hidden_layers": 12,
"type_vocab_size": 2,
"vocab_size": 250112,
"relative_attention_num_buckets": 32,
"attention_scale": false,
"is_dropout": true
}