Training in progress, epoch 1
Browse files- config.json +132 -0
- model.safetensors +3 -0
- runs/Dec26_09-31-34_b4a157e0d31e/events.out.tfevents.1703583106.b4a157e0d31e.1546.7 +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +64 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
@@ -0,0 +1,132 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "cointegrated/rubert-tiny2",
|
3 |
+
"architectures": [
|
4 |
+
"BertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"emb_size": 312,
|
9 |
+
"gradient_checkpointing": false,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.1,
|
12 |
+
"hidden_size": 312,
|
13 |
+
"id2label": {
|
14 |
+
"0": "O",
|
15 |
+
"1": "\u041f\u043e\u0436\u0430\u0440\u044b",
|
16 |
+
"2": "\u0417\u0430\u043a\u043b\u044e\u0447\u0435\u043d\u0438\u0435 \u043f\u043e\u0434 \u0441\u0442\u0440\u0430\u0436\u0443",
|
17 |
+
"3": "\u0412\u0437\u044f\u0442\u043a\u0438",
|
18 |
+
"4": "\u0423\u0433\u043e\u043b\u043e\u0432\u043d\u043e\u0435 \u043e\u0431\u0432\u0438\u043d\u0435\u043d\u0438\u0435",
|
19 |
+
"5": "\u0422\u0435\u0440\u0430\u043a\u0442",
|
20 |
+
"6": "\u042d\u043a\u0441\u0442\u0440\u0435\u043c\u0438\u0437\u043c",
|
21 |
+
"7": "\u041f\u0430\u0434\u0435\u043d\u0438\u0435 \u043f\u0440\u0438\u0431\u044b\u043b\u0438",
|
22 |
+
"8": "\u0410\u0432\u0438\u0430\u043a\u0430\u0442\u0430\u0441\u0442\u0440\u043e\u0444\u0430",
|
23 |
+
"9": "\u0417\u0435\u043c\u043b\u0435\u0442\u0440\u044f\u0441\u0435\u043d\u0438\u0435",
|
24 |
+
"10": "\u041e\u0442\u0437\u044b\u0432 \u043f\u0440\u043e\u0434\u0443\u043a\u0446\u0438\u0438",
|
25 |
+
"11": "\u041a\u043e\u0440\u0440\u0443\u043f\u0446\u0438\u044f",
|
26 |
+
"12": "\u0420\u043e\u0441\u0442 \u0443\u0431\u044b\u0442\u043a\u043e\u0432",
|
27 |
+
"13": "\u0420\u0430\u0441\u0442\u0440\u0430\u0442\u0430 \u0434\u0435\u043d\u0435\u0436\u043d\u044b\u0445 \u0441\u0440\u0435\u0434\u0441\u0442\u0432",
|
28 |
+
"14": "\u0423\u0442\u0435\u0447\u043a\u0430 \u043d\u0435\u0444\u0442\u0438",
|
29 |
+
"15": "\u041a\u043e\u0440\u0430\u0431\u043b\u0435\u043a\u0440\u0443\u0448\u0435\u043d\u0438\u0435",
|
30 |
+
"16": "\u041f\u0440\u0438\u0432\u043b\u0435\u0447\u0435\u043d\u0438\u0435 \u043a \u0443\u0433\u043e\u043b\u043e\u0432\u043d\u043e\u0439 \u043e\u0442\u0432\u0435\u0442\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0441\u0442\u0438",
|
31 |
+
"17": "\u0410\u043a\u0446\u0438\u043e\u043d\u0435\u0440\u043d\u044b\u0439 \u043a\u043e\u043d\u0444\u043b\u0438\u043a\u0442",
|
32 |
+
"18": "\u0418\u0437\u0432\u0435\u0440\u0436\u0435\u043d\u0438\u0435 \u0432\u0443\u043b\u043a\u0430\u043d\u0430",
|
33 |
+
"19": "\u041d\u0430\u0440\u0443\u0448\u0435\u043d\u0438\u0435 \u0437\u0430\u043a\u043e\u043d\u0430",
|
34 |
+
"20": "\u041e\u0431\u0432\u0430\u043b",
|
35 |
+
"21": "\u041f\u0440\u043e\u0442\u0438\u0432\u043e\u0434\u0435\u0439\u0441\u0442\u0432\u0438\u0435 \u043a\u043e\u0440\u0440\u0443\u043f\u0446\u0438\u0438",
|
36 |
+
"22": "\u0417\u0430\u043a\u0440\u044b\u0442\u0438\u0435 \u043f\u0440\u0435\u0434\u043f\u0440\u0438\u044f\u0442\u0438\u044f \u0438\u043b\u0438 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0441\u0442\u0432\u0430",
|
37 |
+
"23": "\u0423\u0431\u0438\u0439\u0441\u0442\u0432\u043e",
|
38 |
+
"24": "\u041d\u0430\u0432\u043e\u0434\u043d\u0435\u043d\u0438\u0435",
|
39 |
+
"25": "\u0423\u0433\u043e\u043b\u043e\u0432\u043d\u043e\u0435 \u0437\u0430\u0434\u0435\u0440\u0436\u0430\u043d\u0438\u0435",
|
40 |
+
"26": "\u0422\u0435\u0445\u043d\u043e\u0433\u0435\u043d\u043d\u044b\u0435 \u043a\u0430\u0442\u0430\u0441\u0442\u0440\u043e\u0444\u044b",
|
41 |
+
"27": "\u0423\u0431\u044b\u0442\u043a\u0438",
|
42 |
+
"28": "\u0421\u0442\u0438\u0445\u0438\u0439\u043d\u044b\u0435 \u0431\u0435\u0434\u0441\u0442\u0432\u0438\u044f",
|
43 |
+
"29": "\u0427\u0440\u0435\u0437\u0432\u044b\u0447\u0430\u0439\u043d\u044b\u0435 \u043f\u0440\u043e\u0438\u0441\u0448\u0435\u0441\u0442\u0432\u0438\u044f, \u0447\u0440\u0435\u0437\u0432\u044b\u0447\u0430\u0439\u043d\u044b\u0435 \u0441\u0438\u0442\u0443\u0430\u0446\u0438\u0438",
|
44 |
+
"30": "\u041f\u043e\u0434\u0434\u0435\u0440\u0436\u043a\u0430 \u0442\u0435\u0440\u0440\u043e\u0440\u0438\u0441\u0442\u043e\u0432",
|
45 |
+
"31": "\u042d\u043a\u043e\u043d\u043e\u043c\u0438\u0447\u0435\u0441\u043a\u0438\u0435 \u043f\u0440\u0435\u0441\u0442\u0443\u043f\u043b\u0435\u043d\u0438\u044f",
|
46 |
+
"32": "\u0424\u0438\u043d\u0430\u043d\u0441\u043e\u0432\u044b\u0435 \u0437\u0430\u0442\u0440\u0443\u0434\u043d\u0435\u043d\u0438\u044f",
|
47 |
+
"33": "\u041f\u0440\u043e\u0431\u043b\u0435\u043c\u044b \u0441 \u0430\u043a\u0442\u0438\u0432\u0430\u043c\u0438",
|
48 |
+
"34": "\u0412\u043e\u0437\u0431\u0443\u0436\u0434\u0435\u043d\u0438\u0435 \u0443\u0433\u043e\u043b\u043e\u0432\u043d\u043e\u0433\u043e \u0434\u0435\u043b\u0430",
|
49 |
+
"35": "\u0417\u0430\u043a\u0440\u044b\u0442\u0438\u0435 \u0444\u0438\u043b\u0438\u0430\u043b\u0430, \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u0438\u0442\u0435\u043b\u044c\u0441\u0442\u0432\u0430",
|
50 |
+
"36": "\u041f\u0430\u0434\u0435\u043d\u0438\u0435 \u043f\u0440\u043e\u0434\u0430\u0436",
|
51 |
+
"37": "\u0416\u0430\u043b\u043e\u0431\u044b \u043f\u043e\u0442\u0440\u0435\u0431\u0438\u0442\u0435\u043b\u0435\u0439",
|
52 |
+
"38": "\u042f\u0434\u0435\u0440\u043d\u044b\u0439 \u0432\u0437\u0440\u044b\u0432",
|
53 |
+
"39": "\u041d\u0430\u0440\u0443\u0448\u0435\u043d\u0438\u0435 \u043e\u0431\u044f\u0437\u0430\u043d\u043d\u043e\u0441\u0442\u0435\u0439",
|
54 |
+
"40": "\u041d\u0430\u0440\u0443\u0448\u0435\u043d\u0438\u0435 \u043f\u0440\u0430\u0432 \u0430\u043a\u0446\u0438\u043e\u043d\u0435\u0440\u043e\u0432",
|
55 |
+
"41": "\u041e\u0442\u043c\u044b\u0432\u0430\u043d\u0438\u0435 \u0434\u0435\u043d\u0435\u0433",
|
56 |
+
"42": "\u041c\u043e\u0448\u0435\u043d\u043d\u0438\u0447\u0435\u0441\u0442\u0432\u043e",
|
57 |
+
"43": "\u0412\u0437\u0440\u044b\u0432 \u0433\u0430\u0437\u0430",
|
58 |
+
"44": "\u0412\u044b\u0435\u043c\u043a\u0438",
|
59 |
+
"45": "\u0411\u0435\u0441\u043f\u043e\u0440\u044f\u0434\u043a\u0438, \u043a\u043e\u043d\u0444\u043b\u0438\u043a\u0442\u044b \u0438 \u0432\u043e\u0439\u043d\u044b",
|
60 |
+
"46": "\u041a\u0430\u0442\u0430\u0441\u0442\u0440\u043e\u0444\u044b, \u043a\u0440\u0443\u0448\u0435\u043d\u0438\u044f",
|
61 |
+
"47": "\u041c\u0430\u043d\u0438\u043f\u0443\u043b\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u0435 \u0440\u044b\u043d\u043a\u043e\u043c",
|
62 |
+
"48": "\u041f\u0430\u0434\u0435\u043d\u0438\u0435 \u043c\u0435\u0442\u0435\u043e\u0440\u0438\u0442\u0430",
|
63 |
+
"49": "\u041d\u0430\u043f\u0430\u0434\u0435\u043d\u0438\u0435"
|
64 |
+
},
|
65 |
+
"initializer_range": 0.02,
|
66 |
+
"intermediate_size": 600,
|
67 |
+
"label2id": {
|
68 |
+
"O": 0,
|
69 |
+
"\u0410\u0432\u0438\u0430\u043a\u0430\u0442\u0430\u0441\u0442\u0440\u043e\u0444\u0430": 8,
|
70 |
+
"\u0410\u043a\u0446\u0438\u043e\u043d\u0435\u0440\u043d\u044b\u0439 \u043a\u043e\u043d\u0444\u043b\u0438\u043a\u0442": 17,
|
71 |
+
"\u0411\u0435\u0441\u043f\u043e\u0440\u044f\u0434\u043a\u0438, \u043a\u043e\u043d\u0444\u043b\u0438\u043a\u0442\u044b \u0438 \u0432\u043e\u0439\u043d\u044b": 45,
|
72 |
+
"\u0412\u0437\u0440\u044b\u0432 \u0433\u0430\u0437\u0430": 43,
|
73 |
+
"\u0412\u0437\u044f\u0442\u043a\u0438": 3,
|
74 |
+
"\u0412\u043e\u0437\u0431\u0443\u0436\u0434\u0435\u043d\u0438\u0435 \u0443\u0433\u043e\u043b\u043e\u0432\u043d\u043e\u0433\u043e \u0434\u0435\u043b\u0430": 34,
|
75 |
+
"\u0412\u044b\u0435\u043c\u043a\u0438": 44,
|
76 |
+
"\u0416\u0430\u043b\u043e\u0431\u044b \u043f\u043e\u0442\u0440\u0435\u0431\u0438\u0442\u0435\u043b\u0435\u0439": 37,
|
77 |
+
"\u0417\u0430\u043a\u043b\u044e\u0447\u0435\u043d\u0438\u0435 \u043f\u043e\u0434 \u0441\u0442\u0440\u0430\u0436\u0443": 2,
|
78 |
+
"\u0417\u0430\u043a\u0440\u044b\u0442\u0438\u0435 \u043f\u0440\u0435\u0434\u043f\u0440\u0438\u044f\u0442\u0438\u044f \u0438\u043b\u0438 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0441\u0442\u0432\u0430": 22,
|
79 |
+
"\u0417\u0430\u043a\u0440\u044b\u0442\u0438\u0435 \u0444\u0438\u043b\u0438\u0430\u043b\u0430, \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u0438\u0442\u0435\u043b\u044c\u0441\u0442\u0432\u0430": 35,
|
80 |
+
"\u0417\u0435\u043c\u043b\u0435\u0442\u0440\u044f\u0441\u0435\u043d\u0438\u0435": 9,
|
81 |
+
"\u0418\u0437\u0432\u0435\u0440\u0436\u0435\u043d\u0438\u0435 \u0432\u0443\u043b\u043a\u0430\u043d\u0430": 18,
|
82 |
+
"\u041a\u0430\u0442\u0430\u0441\u0442\u0440\u043e\u0444\u044b, \u043a\u0440\u0443\u0448\u0435\u043d\u0438\u044f": 46,
|
83 |
+
"\u041a\u043e\u0440\u0430\u0431\u043b\u0435\u043a\u0440\u0443\u0448\u0435\u043d\u0438\u0435": 15,
|
84 |
+
"\u041a\u043e\u0440\u0440\u0443\u043f\u0446\u0438\u044f": 11,
|
85 |
+
"\u041c\u0430\u043d\u0438\u043f\u0443\u043b\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u0435 \u0440\u044b\u043d\u043a\u043e\u043c": 47,
|
86 |
+
"\u041c\u043e\u0448\u0435\u043d\u043d\u0438\u0447\u0435\u0441\u0442\u0432\u043e": 42,
|
87 |
+
"\u041d\u0430\u0432\u043e\u0434\u043d\u0435\u043d\u0438\u0435": 24,
|
88 |
+
"\u041d\u0430\u043f\u0430\u0434\u0435\u043d\u0438\u0435": 49,
|
89 |
+
"\u041d\u0430\u0440\u0443\u0448\u0435\u043d\u0438\u0435 \u0437\u0430\u043a\u043e\u043d\u0430": 19,
|
90 |
+
"\u041d\u0430\u0440\u0443\u0448\u0435\u043d\u0438\u0435 \u043e\u0431\u044f\u0437\u0430\u043d\u043d\u043e\u0441\u0442\u0435\u0439": 39,
|
91 |
+
"\u041d\u0430\u0440\u0443\u0448\u0435\u043d\u0438\u0435 \u043f\u0440\u0430\u0432 \u0430\u043a\u0446\u0438\u043e\u043d\u0435\u0440\u043e\u0432": 40,
|
92 |
+
"\u041e\u0431\u0432\u0430\u043b": 20,
|
93 |
+
"\u041e\u0442\u0437\u044b\u0432 \u043f\u0440\u043e\u0434\u0443\u043a\u0446\u0438\u0438": 10,
|
94 |
+
"\u041e\u0442\u043c\u044b\u0432\u0430\u043d\u0438\u0435 \u0434\u0435\u043d\u0435\u0433": 41,
|
95 |
+
"\u041f\u0430\u0434\u0435\u043d\u0438\u0435 \u043c\u0435\u0442\u0435\u043e\u0440\u0438\u0442\u0430": 48,
|
96 |
+
"\u041f\u0430\u0434\u0435\u043d\u0438\u0435 \u043f\u0440\u0438\u0431\u044b\u043b\u0438": 7,
|
97 |
+
"\u041f\u0430\u0434\u0435\u043d\u0438\u0435 \u043f\u0440\u043e\u0434\u0430\u0436": 36,
|
98 |
+
"\u041f\u043e\u0434\u0434\u0435\u0440\u0436\u043a\u0430 \u0442\u0435\u0440\u0440\u043e\u0440\u0438\u0441\u0442\u043e\u0432": 30,
|
99 |
+
"\u041f\u043e\u0436\u0430\u0440\u044b": 1,
|
100 |
+
"\u041f\u0440\u0438\u0432\u043b\u0435\u0447\u0435\u043d\u0438\u0435 \u043a \u0443\u0433\u043e\u043b\u043e\u0432\u043d\u043e\u0439 \u043e\u0442\u0432\u0435\u0442\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0441\u0442\u0438": 16,
|
101 |
+
"\u041f\u0440\u043e\u0431\u043b\u0435\u043c\u044b \u0441 \u0430\u043a\u0442\u0438\u0432\u0430\u043c\u0438": 33,
|
102 |
+
"\u041f\u0440\u043e\u0442\u0438\u0432\u043e\u0434\u0435\u0439\u0441\u0442\u0432\u0438\u0435 \u043a\u043e\u0440\u0440\u0443\u043f\u0446\u0438\u0438": 21,
|
103 |
+
"\u0420\u0430\u0441\u0442\u0440\u0430\u0442\u0430 \u0434\u0435\u043d\u0435\u0436\u043d\u044b\u0445 \u0441\u0440\u0435\u0434\u0441\u0442\u0432": 13,
|
104 |
+
"\u0420\u043e\u0441\u0442 \u0443\u0431\u044b\u0442\u043a\u043e\u0432": 12,
|
105 |
+
"\u0421\u0442\u0438\u0445\u0438\u0439\u043d\u044b\u0435 \u0431\u0435\u0434\u0441\u0442\u0432\u0438\u044f": 28,
|
106 |
+
"\u0422\u0435\u0440\u0430\u043a\u0442": 5,
|
107 |
+
"\u0422\u0435\u0445\u043d\u043e\u0433\u0435\u043d\u043d\u044b\u0435 \u043a\u0430\u0442\u0430\u0441\u0442\u0440\u043e\u0444\u044b": 26,
|
108 |
+
"\u0423\u0431\u0438\u0439\u0441\u0442\u0432\u043e": 23,
|
109 |
+
"\u0423\u0431\u044b\u0442\u043a\u0438": 27,
|
110 |
+
"\u0423\u0433\u043e\u043b\u043e\u0432\u043d\u043e\u0435 \u0437\u0430\u0434\u0435\u0440\u0436\u0430\u043d\u0438\u0435": 25,
|
111 |
+
"\u0423\u0433\u043e\u043b\u043e\u0432\u043d\u043e\u0435 \u043e\u0431\u0432\u0438\u043d\u0435\u043d\u0438\u0435": 4,
|
112 |
+
"\u0423\u0442\u0435\u0447\u043a\u0430 \u043d\u0435\u0444\u0442\u0438": 14,
|
113 |
+
"\u0424\u0438\u043d\u0430\u043d\u0441\u043e\u0432\u044b\u0435 \u0437\u0430\u0442\u0440\u0443\u0434\u043d\u0435\u043d\u0438\u044f": 32,
|
114 |
+
"\u0427\u0440\u0435\u0437\u0432\u044b\u0447\u0430\u0439\u043d\u044b\u0435 \u043f\u0440\u043e\u0438\u0441\u0448\u0435\u0441\u0442\u0432\u0438\u044f, \u0447\u0440\u0435\u0437\u0432\u044b\u0447\u0430\u0439\u043d\u044b\u0435 \u0441\u0438\u0442\u0443\u0430\u0446\u0438\u0438": 29,
|
115 |
+
"\u042d\u043a\u043e\u043d\u043e\u043c\u0438\u0447\u0435\u0441\u043a\u0438\u0435 \u043f\u0440\u0435\u0441\u0442\u0443\u043f\u043b\u0435\u043d\u0438\u044f": 31,
|
116 |
+
"\u042d\u043a\u0441\u0442\u0440\u0435\u043c\u0438\u0437\u043c": 6,
|
117 |
+
"\u042f\u0434\u0435\u0440\u043d\u044b\u0439 \u0432\u0437\u0440\u044b\u0432": 38
|
118 |
+
},
|
119 |
+
"layer_norm_eps": 1e-12,
|
120 |
+
"max_position_embeddings": 2048,
|
121 |
+
"model_type": "bert",
|
122 |
+
"num_attention_heads": 12,
|
123 |
+
"num_hidden_layers": 3,
|
124 |
+
"pad_token_id": 0,
|
125 |
+
"position_embedding_type": "absolute",
|
126 |
+
"problem_type": "single_label_classification",
|
127 |
+
"torch_dtype": "float32",
|
128 |
+
"transformers_version": "4.35.2",
|
129 |
+
"type_vocab_size": 2,
|
130 |
+
"use_cache": true,
|
131 |
+
"vocab_size": 83828
|
132 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:304960eea1a45a336acdafa4c2a8de8d7e75619e4ae69f73c1280d31e20ecb14
|
3 |
+
size 116844232
|
runs/Dec26_09-31-34_b4a157e0d31e/events.out.tfevents.1703583106.b4a157e0d31e.1546.7
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6c2bd952e1cd65e9a4c22b3d3cbbeec0be615e06f3ac4b754cf7775819384a0d
|
3 |
+
size 16034
|
special_tokens_map.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"sep_token": "[SEP]",
|
6 |
+
"unk_token": "[UNK]"
|
7 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "[UNK]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"4": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"clean_up_tokenization_spaces": true,
|
45 |
+
"cls_token": "[CLS]",
|
46 |
+
"do_basic_tokenize": true,
|
47 |
+
"do_lower_case": false,
|
48 |
+
"mask_token": "[MASK]",
|
49 |
+
"max_length": 512,
|
50 |
+
"model_max_length": 2048,
|
51 |
+
"never_split": null,
|
52 |
+
"pad_to_multiple_of": null,
|
53 |
+
"pad_token": "[PAD]",
|
54 |
+
"pad_token_type_id": 0,
|
55 |
+
"padding_side": "right",
|
56 |
+
"sep_token": "[SEP]",
|
57 |
+
"stride": 0,
|
58 |
+
"strip_accents": null,
|
59 |
+
"tokenize_chinese_chars": true,
|
60 |
+
"tokenizer_class": "BertTokenizer",
|
61 |
+
"truncation_side": "right",
|
62 |
+
"truncation_strategy": "longest_first",
|
63 |
+
"unk_token": "[UNK]"
|
64 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ab9d45adf38d0f2f2fd94702c756aba077d71afcb37bdb5e4a881d029321a79e
|
3 |
+
size 4600
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|