qwazer's picture
Training in progress, epoch 1
e9ade1e verified
raw
history blame
1.23 kB
{
"_name_or_path": "cointegrated/rubert-tiny2",
"architectures": [
"BertForTokenClassification"
],
"attention_probs_dropout_prob": 0.1,
"classifier_dropout": null,
"emb_size": 312,
"gradient_checkpointing": false,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 312,
"id2label": {
"0": "OTHER",
"1": "REGION",
"2": "REGION_TYPE",
"3": "AREA",
"4": "AREA_TYPE",
"5": "TERRITORY",
"6": "TERRITORY_TYPE",
"7": "CITY",
"8": "CITY_TYPE",
"9": "STREET",
"10": "STREET_TYPE",
"11": "DELIMITER"
},
"initializer_range": 0.02,
"intermediate_size": 600,
"label2id": {
"AREA": 3,
"AREA_TYPE": 4,
"CITY": 7,
"CITY_TYPE": 8,
"DELIMITER": 11,
"OTHER": 0,
"REGION": 1,
"REGION_TYPE": 2,
"STREET": 9,
"STREET_TYPE": 10,
"TERRITORY": 5,
"TERRITORY_TYPE": 6
},
"layer_norm_eps": 1e-12,
"max_position_embeddings": 2048,
"model_type": "bert",
"num_attention_heads": 12,
"num_hidden_layers": 3,
"pad_token_id": 0,
"position_embedding_type": "absolute",
"torch_dtype": "float32",
"transformers_version": "4.40.2",
"type_vocab_size": 2,
"use_cache": true,
"vocab_size": 83828
}