pa-shk's picture
Training in progress, epoch 1
bfe2062
{
"_name_or_path": "DeepPavlov/distilrubert-tiny-cased-conversational-v1",
"activation": "gelu",
"architectures": [
"DistilBertForSequenceClassification"
],
"attention_dropout": 0.1,
"dim": 264,
"dropout": 0.1,
"hidden_dim": 792,
"id2label": {
"0": "\u0431\u043b\u0430\u0433\u043e\u0434\u0430\u0440\u043d\u043e\u0441\u0442\u044c \u0432 \u0442\u0435\u043a\u0441\u0442\u0435",
"1": "\u0432 \u043f\u0443\u0442\u0438",
"2": "\u0432\u043e\u0441\u044c\u043c\u043e\u0435 \u043c\u0430\u0440\u0442\u0430",
"3": "\u0434\u0435\u043d\u044c \u0440\u043e\u0436\u0434\u0435\u043d\u0438\u044f",
"4": "\u0436\u0438\u0442\u0435\u0439\u0441\u043a\u043e\u0435",
"5": "\u0438\u0437 \u043f\u0443\u0442\u0435\u0448\u0435\u0441\u0442\u0432\u0438\u044f",
"6": "\u0438\u043c\u0435\u043d\u0438\u043d\u044b",
"7": "\u043d\u043e\u0432\u043e\u0441\u0442\u0438",
"8": "\u043d\u043e\u0432\u044b\u0439 \u0433\u043e\u0434",
"9": "\u043e\u043a\u0442\u044f\u0431\u0440\u044c\u0441\u043a\u0430\u044f \u0440\u0435\u0432\u043e\u043b\u044e\u0446\u0438\u044f",
"10": "\u043e\u0442\u0434\u044b\u0445",
"11": "\u043f\u0430\u0441\u0445\u0430",
"12": "\u043f\u0435\u0440\u0432\u043e\u0435 \u043c\u0430\u044f",
"13": "\u043f\u0438\u0441\u044c\u043c\u0430 \u043e\u0442 \u0441\u0435\u043c\u044c\u0438",
"14": "\u043f\u0438\u0441\u044c\u043c\u0430 \u0441\u0435\u043c\u044c\u0435",
"15": "\u043f\u043e\u0433\u043e\u0434\u0430",
"16": "\u043f\u043e\u0435\u0437\u0434\u043a\u0430",
"17": "\u043f\u043e\u0437\u0434\u0440\u0430\u0432\u043b\u0435\u043d\u0438\u0435 \u0432 \u0442\u0435\u043a\u0441\u0442\u0435",
"18": "\u043f\u043e\u0446\u0435\u043b\u0443\u0438 \u0438\u0437 \u043f\u0438\u0441\u0435\u043c",
"19": "\u043f\u0440\u0438\u0433\u043b\u0430\u0448\u0435\u043d\u0438\u0435 \u0432 \u0442\u0435\u043a\u0441\u0442\u0435",
"20": "\u043f\u0440\u043e\u0441\u044c\u0431\u0430",
"21": "\u0440\u043e\u0436\u0434\u0435\u0441\u0442\u0432\u043e",
"22": "\u0441\u043e\u0446\u0438\u0430\u043b\u044c\u043d\u043e\u0435",
"23": "\u0443\u0447\u0435\u0431\u0430",
"24": "\u0448\u043b\u044e \u043f\u0440\u0438\u0432\u0435\u0442"
},
"initializer_range": 0.02,
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1,
"LABEL_10": 10,
"LABEL_11": 11,
"LABEL_12": 12,
"LABEL_13": 13,
"LABEL_14": 14,
"LABEL_15": 15,
"LABEL_16": 16,
"LABEL_17": 17,
"LABEL_18": 18,
"LABEL_19": 19,
"LABEL_2": 2,
"LABEL_20": 20,
"LABEL_21": 21,
"LABEL_22": 22,
"LABEL_23": 23,
"LABEL_24": 24,
"LABEL_3": 3,
"LABEL_4": 4,
"LABEL_5": 5,
"LABEL_6": 6,
"LABEL_7": 7,
"LABEL_8": 8,
"LABEL_9": 9
},
"max_position_embeddings": 512,
"model_type": "distilbert",
"n_heads": 12,
"n_layers": 3,
"pad_token_id": 0,
"problem_type": "multi_label_classification",
"qa_dropout": 0.1,
"seq_classif_dropout": 0.2,
"sinusoidal_pos_embds": false,
"torch_dtype": "float32",
"transformers_version": "4.28.0",
"vocab_size": 30522
}