2023MLMA_LAB9_task2 / config.json
yujie07's picture
Training in progress, epoch 1
bb90677
{
"_name_or_path": "microsoft/biogpt",
"activation_dropout": 0.0,
"activation_function": "gelu_new",
"architectures": [
"GPT2ForTokenClassification"
],
"attention_probs_dropout_prob": 0.1,
"attn_pdrop": 0.1,
"bos_token_id": 0,
"embd_pdrop": 0.1,
"eos_token_id": 2,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"id2label": {
"0": "O",
"1": "B-Disease",
"2": "I-Disease"
},
"initializer_range": 0.02,
"intermediate_size": 4096,
"label2id": {
"B-Disease": 1,
"I-Disease": 2,
"O": 0
},
"layer_norm_eps": 1e-12,
"layer_norm_epsilon": 1e-05,
"layerdrop": 0.0,
"model_type": "gpt2",
"n_embd": 1024,
"n_head": 16,
"n_inner": null,
"n_layer": 24,
"n_positions": 1024,
"pad_token_id": 1,
"reorder_and_upcast_attn": false,
"resid_pdrop": 0.1,
"scale_attn_by_inverse_layer_idx": false,
"scale_attn_weights": true,
"scale_embedding": true,
"summary_activation": null,
"summary_first_dropout": 0.1,
"summary_proj_to_labels": true,
"summary_type": "cls_index",
"summary_use_proj": true,
"torch_dtype": "float32",
"transformers_version": "4.28.1",
"use_cache": true,
"vocab_size": 42384
}