bort / config.json
{
"architectures": [
"BertForMaskedLM"
],
"attention_probs_dropout_prob": 0.1,
"gradient_checkpointing": false,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 1024,
"initializer_range": 0.02,
"intermediate_size": 768,
"layer_norm_eps": 1e-05,
"max_position_embeddings": 512,
"model_type": "bert",
"num_attention_heads": 8,
"num_hidden_layers": 4,
"pad_token_id": 1,
"tokenizer_class": "RobertaTokenizer",
"type_vocab_size": 1,
"vocab_size": 50265
}