adsplus / roberta /mt_roberta_v3_5_config.json
fuliucansheng's picture
add mt roberta v3.5 config & vocab & baseline weight
f64578d
{
"attention_probs_dropout_prob": 0.1,
"finetuning_task": null,
"bos_token_id": 0,
"eos_token_id": 2,
"pad_token_id": 1,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 1024,
"initializer_range": 0.02,
"intermediate_size": 4096,
"layer_norm_eps": 1e-05,
"max_position_embeddings": 514,
"num_attention_heads": 16,
"num_hidden_layers": 36,
"num_labels": 2,
"output_attentions": false,
"output_hidden_states": false,
"torchscript": false,
"type_vocab_size": 2,
"vocab_size": 50265
}