roberta-large-aces / config.json
gijs's picture
Training in progress, epoch 1
a92b7ce
{
"_name_or_path": "roberta-large",
"architectures": [
"RobertaForTokenClassification"
],
"attention_probs_dropout_prob": 0.1,
"bos_token_id": 0,
"classifier_dropout": null,
"eos_token_id": 2,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 1024,
"id2label": {
"0": "O",
"1": "WHO",
"2": "WHAT",
"3": "HOW",
"4": "WHEN",
"5": "WHERE",
"6": "SIGNAL",
"7": "AUDITORY_SENSATION",
"8": "NON_AUDITORY_SENSATION",
"9": "OTHER"
},
"initializer_range": 0.02,
"intermediate_size": 4096,
"label2id": {
"AUDITORY_SENSATION": 7,
"HOW": 3,
"NON_AUDITORY_SENSATION": 8,
"O": 0,
"OTHER": 9,
"SIGNAL": 6,
"WHAT": 2,
"WHEN": 4,
"WHERE": 5,
"WHO": 1
},
"layer_norm_eps": 1e-05,
"max_position_embeddings": 514,
"model_type": "roberta",
"num_attention_heads": 16,
"num_hidden_layers": 24,
"pad_token_id": 1,
"position_embedding_type": "absolute",
"torch_dtype": "float32",
"transformers_version": "4.25.1",
"type_vocab_size": 1,
"use_cache": true,
"vocab_size": 50265
}