Biu3010 commited on
Commit
2a1e03a
·
verified ·
1 Parent(s): 72d8a04

Upload 8 files

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
config.json CHANGED
@@ -1,31 +1,57 @@
1
  {
 
 
 
 
 
2
  "architectures": [
3
- "T5ForConditionalGeneration"
4
  ],
 
 
 
5
  "classifier_dropout": 0.0,
6
- "d_ff": 3072,
7
- "d_kv": 64,
8
- "d_model": 768,
9
- "decoder_start_token_id": 0,
10
- "dense_act_fn": "relu",
11
- "dropout_rate": 0.1,
 
12
  "dtype": "float32",
13
- "eos_token_id": 1,
14
- "feed_forward_proj": "relu",
15
- "initializer_factor": 1.0,
 
 
 
 
 
 
 
 
 
 
 
16
  "is_encoder_decoder": true,
17
- "is_gated_act": false,
18
- "layer_norm_epsilon": 1e-06,
19
- "model_type": "t5",
20
- "n_positions": 512,
21
- "num_decoder_layers": 12,
22
- "num_heads": 12,
23
- "num_layers": 12,
 
 
 
 
 
24
  "output_past": true,
25
- "pad_token_id": 0,
26
- "relative_attention_max_distance": 128,
27
- "relative_attention_num_buckets": 32,
28
- "transformers_version": "4.57.1",
 
29
  "use_cache": true,
30
- "vocab_size": 36096
31
  }
 
1
  {
2
+ "_num_labels": 3,
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": true,
7
  "architectures": [
8
+ "MBartForConditionalGeneration"
9
  ],
10
+ "attention_dropout": 0.0,
11
+ "bos_token_id": 0,
12
+ "classif_dropout": 0.0,
13
  "classifier_dropout": 0.0,
14
+ "d_model": 1024,
15
+ "decoder_attention_heads": 16,
16
+ "decoder_ffn_dim": 4096,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 12,
19
+ "decoder_start_token_id": 2,
20
+ "dropout": 0.1,
21
  "dtype": "float32",
22
+ "early_stopping": null,
23
+ "encoder_attention_heads": 16,
24
+ "encoder_ffn_dim": 4096,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 12,
27
+ "eos_token_id": 2,
28
+ "forced_eos_token_id": 2,
29
+ "gradient_checkpointing": false,
30
+ "id2label": {
31
+ "0": "LABEL_0",
32
+ "1": "LABEL_1",
33
+ "2": "LABEL_2"
34
+ },
35
+ "init_std": 0.02,
36
  "is_encoder_decoder": true,
37
+ "label2id": {
38
+ "LABEL_0": 0,
39
+ "LABEL_1": 1,
40
+ "LABEL_2": 2
41
+ },
42
+ "max_length": null,
43
+ "max_position_embeddings": 1024,
44
+ "model_type": "mbart",
45
+ "normalize_before": true,
46
+ "normalize_embedding": true,
47
+ "num_beams": null,
48
+ "num_hidden_layers": 12,
49
  "output_past": true,
50
+ "pad_token_id": 1,
51
+ "scale_embedding": true,
52
+ "static_position_embeddings": false,
53
+ "tokenizer_class": "MBart50Tokenizer",
54
+ "transformers_version": "4.57.3",
55
  "use_cache": true,
56
+ "vocab_size": 250054
57
  }
generation_config.json CHANGED
@@ -1,8 +1,14 @@
1
  {
2
- "decoder_start_token_id": 0,
 
 
 
3
  "eos_token_id": [
4
- 1
5
  ],
6
- "pad_token_id": 0,
7
- "transformers_version": "4.57.1"
 
 
 
8
  }
 
1
  {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "decoder_start_token_id": 2,
5
+ "early_stopping": true,
6
  "eos_token_id": [
7
+ 2
8
  ],
9
+ "forced_eos_token_id": 2,
10
+ "max_length": 200,
11
+ "num_beams": 5,
12
+ "pad_token_id": 1,
13
+ "transformers_version": "4.57.3"
14
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b8d869602470954b21fd4befc598d3b4c0fe17da09f50ad76934a896bf6936ce
3
- size 903834408
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3bfbedfdd3d64f7ce1aea5a02293d65a54daa7cfe639c5b0c6faa5f08bb8391
3
+ size 2444578688
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
special_tokens_map.json CHANGED
@@ -1,103 +1,105 @@
1
  {
2
  "additional_special_tokens": [
3
- "<extra_id_0>",
4
- "<extra_id_1>",
5
- "<extra_id_2>",
6
- "<extra_id_3>",
7
- "<extra_id_4>",
8
- "<extra_id_5>",
9
- "<extra_id_6>",
10
- "<extra_id_7>",
11
- "<extra_id_8>",
12
- "<extra_id_9>",
13
- "<extra_id_10>",
14
- "<extra_id_11>",
15
- "<extra_id_12>",
16
- "<extra_id_13>",
17
- "<extra_id_14>",
18
- "<extra_id_15>",
19
- "<extra_id_16>",
20
- "<extra_id_17>",
21
- "<extra_id_18>",
22
- "<extra_id_19>",
23
- "<extra_id_20>",
24
- "<extra_id_21>",
25
- "<extra_id_22>",
26
- "<extra_id_23>",
27
- "<extra_id_24>",
28
- "<extra_id_25>",
29
- "<extra_id_26>",
30
- "<extra_id_27>",
31
- "<extra_id_28>",
32
- "<extra_id_29>",
33
- "<extra_id_30>",
34
- "<extra_id_31>",
35
- "<extra_id_32>",
36
- "<extra_id_33>",
37
- "<extra_id_34>",
38
- "<extra_id_35>",
39
- "<extra_id_36>",
40
- "<extra_id_37>",
41
- "<extra_id_38>",
42
- "<extra_id_39>",
43
- "<extra_id_40>",
44
- "<extra_id_41>",
45
- "<extra_id_42>",
46
- "<extra_id_43>",
47
- "<extra_id_44>",
48
- "<extra_id_45>",
49
- "<extra_id_46>",
50
- "<extra_id_47>",
51
- "<extra_id_48>",
52
- "<extra_id_49>",
53
- "<extra_id_50>",
54
- "<extra_id_51>",
55
- "<extra_id_52>",
56
- "<extra_id_53>",
57
- "<extra_id_54>",
58
- "<extra_id_55>",
59
- "<extra_id_56>",
60
- "<extra_id_57>",
61
- "<extra_id_58>",
62
- "<extra_id_59>",
63
- "<extra_id_60>",
64
- "<extra_id_61>",
65
- "<extra_id_62>",
66
- "<extra_id_63>",
67
- "<extra_id_64>",
68
- "<extra_id_65>",
69
- "<extra_id_66>",
70
- "<extra_id_67>",
71
- "<extra_id_68>",
72
- "<extra_id_69>",
73
- "<extra_id_70>",
74
- "<extra_id_71>",
75
- "<extra_id_72>",
76
- "<extra_id_73>",
77
- "<extra_id_74>",
78
- "<extra_id_75>",
79
- "<extra_id_76>",
80
- "<extra_id_77>",
81
- "<extra_id_78>",
82
- "<extra_id_79>",
83
- "<extra_id_80>",
84
- "<extra_id_81>",
85
- "<extra_id_82>",
86
- "<extra_id_83>",
87
- "<extra_id_84>",
88
- "<extra_id_85>",
89
- "<extra_id_86>",
90
- "<extra_id_87>",
91
- "<extra_id_88>",
92
- "<extra_id_89>",
93
- "<extra_id_90>",
94
- "<extra_id_91>",
95
- "<extra_id_92>",
96
- "<extra_id_93>",
97
- "<extra_id_94>",
98
- "<extra_id_95>"
99
  ],
100
- "eos_token": "</s>",
101
- "pad_token": "<pad>",
102
- "unk_token": "<unk>"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
103
  }
 
1
  {
2
  "additional_special_tokens": [
3
+ "ar_AR",
4
+ "cs_CZ",
5
+ "de_DE",
6
+ "en_XX",
7
+ "es_XX",
8
+ "et_EE",
9
+ "fi_FI",
10
+ "fr_XX",
11
+ "gu_IN",
12
+ "hi_IN",
13
+ "it_IT",
14
+ "ja_XX",
15
+ "kk_KZ",
16
+ "ko_KR",
17
+ "lt_LT",
18
+ "lv_LV",
19
+ "my_MM",
20
+ "ne_NP",
21
+ "nl_XX",
22
+ "ro_RO",
23
+ "ru_RU",
24
+ "si_LK",
25
+ "tr_TR",
26
+ "vi_VN",
27
+ "zh_CN",
28
+ "af_ZA",
29
+ "az_AZ",
30
+ "bn_IN",
31
+ "fa_IR",
32
+ "he_IL",
33
+ "hr_HR",
34
+ "id_ID",
35
+ "ka_GE",
36
+ "km_KH",
37
+ "mk_MK",
38
+ "ml_IN",
39
+ "mn_MN",
40
+ "mr_IN",
41
+ "pl_PL",
42
+ "ps_AF",
43
+ "pt_XX",
44
+ "sv_SE",
45
+ "sw_KE",
46
+ "ta_IN",
47
+ "te_IN",
48
+ "th_TH",
49
+ "tl_XX",
50
+ "uk_UA",
51
+ "ur_PK",
52
+ "xh_ZA",
53
+ "gl_ES",
54
+ "sl_SI"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
55
  ],
56
+ "bos_token": {
57
+ "content": "<s>",
58
+ "lstrip": false,
59
+ "normalized": false,
60
+ "rstrip": false,
61
+ "single_word": false
62
+ },
63
+ "cls_token": {
64
+ "content": "<s>",
65
+ "lstrip": false,
66
+ "normalized": false,
67
+ "rstrip": false,
68
+ "single_word": false
69
+ },
70
+ "eos_token": {
71
+ "content": "</s>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false
76
+ },
77
+ "mask_token": {
78
+ "content": "<mask>",
79
+ "lstrip": true,
80
+ "normalized": true,
81
+ "rstrip": false,
82
+ "single_word": false
83
+ },
84
+ "pad_token": {
85
+ "content": "<pad>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false
90
+ },
91
+ "sep_token": {
92
+ "content": "</s>",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false
97
+ },
98
+ "unk_token": {
99
+ "content": "<unk>",
100
+ "lstrip": false,
101
+ "normalized": false,
102
+ "rstrip": false,
103
+ "single_word": false
104
+ }
105
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -1,8 +1,7 @@
1
  {
2
- "add_prefix_space": null,
3
  "added_tokens_decoder": {
4
  "0": {
5
- "content": "<pad>",
6
  "lstrip": false,
7
  "normalized": false,
8
  "rstrip": false,
@@ -10,7 +9,7 @@
10
  "special": true
11
  },
12
  "1": {
13
- "content": "</s>",
14
  "lstrip": false,
15
  "normalized": false,
16
  "rstrip": false,
@@ -18,887 +17,519 @@
18
  "special": true
19
  },
20
  "2": {
21
- "content": "<unk>",
22
- "lstrip": false,
23
- "normalized": false,
24
- "rstrip": false,
25
- "single_word": false,
26
- "special": true
27
- },
28
- "36000": {
29
- "content": "<extra_id_95>",
30
- "lstrip": false,
31
- "normalized": false,
32
- "rstrip": false,
33
- "single_word": false,
34
- "special": true
35
- },
36
- "36001": {
37
- "content": "<extra_id_94>",
38
- "lstrip": false,
39
- "normalized": false,
40
- "rstrip": false,
41
- "single_word": false,
42
- "special": true
43
- },
44
- "36002": {
45
- "content": "<extra_id_93>",
46
- "lstrip": false,
47
- "normalized": false,
48
- "rstrip": false,
49
- "single_word": false,
50
- "special": true
51
- },
52
- "36003": {
53
- "content": "<extra_id_92>",
54
- "lstrip": false,
55
- "normalized": false,
56
- "rstrip": false,
57
- "single_word": false,
58
- "special": true
59
- },
60
- "36004": {
61
- "content": "<extra_id_91>",
62
- "lstrip": false,
63
- "normalized": false,
64
- "rstrip": false,
65
- "single_word": false,
66
- "special": true
67
- },
68
- "36005": {
69
- "content": "<extra_id_90>",
70
- "lstrip": false,
71
- "normalized": false,
72
- "rstrip": false,
73
- "single_word": false,
74
- "special": true
75
- },
76
- "36006": {
77
- "content": "<extra_id_89>",
78
- "lstrip": false,
79
- "normalized": false,
80
- "rstrip": false,
81
- "single_word": false,
82
- "special": true
83
- },
84
- "36007": {
85
- "content": "<extra_id_88>",
86
- "lstrip": false,
87
- "normalized": false,
88
- "rstrip": false,
89
- "single_word": false,
90
- "special": true
91
- },
92
- "36008": {
93
- "content": "<extra_id_87>",
94
- "lstrip": false,
95
- "normalized": false,
96
- "rstrip": false,
97
- "single_word": false,
98
- "special": true
99
- },
100
- "36009": {
101
- "content": "<extra_id_86>",
102
- "lstrip": false,
103
- "normalized": false,
104
- "rstrip": false,
105
- "single_word": false,
106
- "special": true
107
- },
108
- "36010": {
109
- "content": "<extra_id_85>",
110
- "lstrip": false,
111
- "normalized": false,
112
- "rstrip": false,
113
- "single_word": false,
114
- "special": true
115
- },
116
- "36011": {
117
- "content": "<extra_id_84>",
118
- "lstrip": false,
119
- "normalized": false,
120
- "rstrip": false,
121
- "single_word": false,
122
- "special": true
123
- },
124
- "36012": {
125
- "content": "<extra_id_83>",
126
- "lstrip": false,
127
- "normalized": false,
128
- "rstrip": false,
129
- "single_word": false,
130
- "special": true
131
- },
132
- "36013": {
133
- "content": "<extra_id_82>",
134
- "lstrip": false,
135
- "normalized": false,
136
- "rstrip": false,
137
- "single_word": false,
138
- "special": true
139
- },
140
- "36014": {
141
- "content": "<extra_id_81>",
142
- "lstrip": false,
143
- "normalized": false,
144
- "rstrip": false,
145
- "single_word": false,
146
- "special": true
147
- },
148
- "36015": {
149
- "content": "<extra_id_80>",
150
- "lstrip": false,
151
- "normalized": false,
152
- "rstrip": false,
153
- "single_word": false,
154
- "special": true
155
- },
156
- "36016": {
157
- "content": "<extra_id_79>",
158
- "lstrip": false,
159
- "normalized": false,
160
- "rstrip": false,
161
- "single_word": false,
162
- "special": true
163
- },
164
- "36017": {
165
- "content": "<extra_id_78>",
166
- "lstrip": false,
167
- "normalized": false,
168
- "rstrip": false,
169
- "single_word": false,
170
- "special": true
171
- },
172
- "36018": {
173
- "content": "<extra_id_77>",
174
- "lstrip": false,
175
- "normalized": false,
176
- "rstrip": false,
177
- "single_word": false,
178
- "special": true
179
- },
180
- "36019": {
181
- "content": "<extra_id_76>",
182
- "lstrip": false,
183
- "normalized": false,
184
- "rstrip": false,
185
- "single_word": false,
186
- "special": true
187
- },
188
- "36020": {
189
- "content": "<extra_id_75>",
190
- "lstrip": false,
191
- "normalized": false,
192
- "rstrip": false,
193
- "single_word": false,
194
- "special": true
195
- },
196
- "36021": {
197
- "content": "<extra_id_74>",
198
- "lstrip": false,
199
- "normalized": false,
200
- "rstrip": false,
201
- "single_word": false,
202
- "special": true
203
- },
204
- "36022": {
205
- "content": "<extra_id_73>",
206
- "lstrip": false,
207
- "normalized": false,
208
- "rstrip": false,
209
- "single_word": false,
210
- "special": true
211
- },
212
- "36023": {
213
- "content": "<extra_id_72>",
214
- "lstrip": false,
215
- "normalized": false,
216
- "rstrip": false,
217
- "single_word": false,
218
- "special": true
219
- },
220
- "36024": {
221
- "content": "<extra_id_71>",
222
- "lstrip": false,
223
- "normalized": false,
224
- "rstrip": false,
225
- "single_word": false,
226
- "special": true
227
- },
228
- "36025": {
229
- "content": "<extra_id_70>",
230
- "lstrip": false,
231
- "normalized": false,
232
- "rstrip": false,
233
- "single_word": false,
234
- "special": true
235
- },
236
- "36026": {
237
- "content": "<extra_id_69>",
238
- "lstrip": false,
239
- "normalized": false,
240
- "rstrip": false,
241
- "single_word": false,
242
- "special": true
243
- },
244
- "36027": {
245
- "content": "<extra_id_68>",
246
- "lstrip": false,
247
- "normalized": false,
248
- "rstrip": false,
249
- "single_word": false,
250
- "special": true
251
- },
252
- "36028": {
253
- "content": "<extra_id_67>",
254
- "lstrip": false,
255
- "normalized": false,
256
- "rstrip": false,
257
- "single_word": false,
258
- "special": true
259
- },
260
- "36029": {
261
- "content": "<extra_id_66>",
262
- "lstrip": false,
263
- "normalized": false,
264
- "rstrip": false,
265
- "single_word": false,
266
- "special": true
267
- },
268
- "36030": {
269
- "content": "<extra_id_65>",
270
- "lstrip": false,
271
- "normalized": false,
272
- "rstrip": false,
273
- "single_word": false,
274
- "special": true
275
- },
276
- "36031": {
277
- "content": "<extra_id_64>",
278
- "lstrip": false,
279
- "normalized": false,
280
- "rstrip": false,
281
- "single_word": false,
282
- "special": true
283
- },
284
- "36032": {
285
- "content": "<extra_id_63>",
286
- "lstrip": false,
287
- "normalized": false,
288
- "rstrip": false,
289
- "single_word": false,
290
- "special": true
291
- },
292
- "36033": {
293
- "content": "<extra_id_62>",
294
- "lstrip": false,
295
- "normalized": false,
296
- "rstrip": false,
297
- "single_word": false,
298
- "special": true
299
- },
300
- "36034": {
301
- "content": "<extra_id_61>",
302
- "lstrip": false,
303
- "normalized": false,
304
- "rstrip": false,
305
- "single_word": false,
306
- "special": true
307
- },
308
- "36035": {
309
- "content": "<extra_id_60>",
310
- "lstrip": false,
311
- "normalized": false,
312
- "rstrip": false,
313
- "single_word": false,
314
- "special": true
315
- },
316
- "36036": {
317
- "content": "<extra_id_59>",
318
- "lstrip": false,
319
- "normalized": false,
320
- "rstrip": false,
321
- "single_word": false,
322
- "special": true
323
- },
324
- "36037": {
325
- "content": "<extra_id_58>",
326
- "lstrip": false,
327
- "normalized": false,
328
- "rstrip": false,
329
- "single_word": false,
330
- "special": true
331
- },
332
- "36038": {
333
- "content": "<extra_id_57>",
334
- "lstrip": false,
335
- "normalized": false,
336
- "rstrip": false,
337
- "single_word": false,
338
- "special": true
339
- },
340
- "36039": {
341
- "content": "<extra_id_56>",
342
- "lstrip": false,
343
- "normalized": false,
344
- "rstrip": false,
345
- "single_word": false,
346
- "special": true
347
- },
348
- "36040": {
349
- "content": "<extra_id_55>",
350
- "lstrip": false,
351
- "normalized": false,
352
- "rstrip": false,
353
- "single_word": false,
354
- "special": true
355
- },
356
- "36041": {
357
- "content": "<extra_id_54>",
358
  "lstrip": false,
359
  "normalized": false,
360
  "rstrip": false,
361
  "single_word": false,
362
  "special": true
363
  },
364
- "36042": {
365
- "content": "<extra_id_53>",
366
  "lstrip": false,
367
  "normalized": false,
368
  "rstrip": false,
369
  "single_word": false,
370
  "special": true
371
  },
372
- "36043": {
373
- "content": "<extra_id_52>",
374
  "lstrip": false,
375
  "normalized": false,
376
  "rstrip": false,
377
  "single_word": false,
378
  "special": true
379
  },
380
- "36044": {
381
- "content": "<extra_id_51>",
382
  "lstrip": false,
383
  "normalized": false,
384
  "rstrip": false,
385
  "single_word": false,
386
  "special": true
387
  },
388
- "36045": {
389
- "content": "<extra_id_50>",
390
  "lstrip": false,
391
  "normalized": false,
392
  "rstrip": false,
393
  "single_word": false,
394
  "special": true
395
  },
396
- "36046": {
397
- "content": "<extra_id_49>",
398
  "lstrip": false,
399
  "normalized": false,
400
  "rstrip": false,
401
  "single_word": false,
402
  "special": true
403
  },
404
- "36047": {
405
- "content": "<extra_id_48>",
406
  "lstrip": false,
407
  "normalized": false,
408
  "rstrip": false,
409
  "single_word": false,
410
  "special": true
411
  },
412
- "36048": {
413
- "content": "<extra_id_47>",
414
  "lstrip": false,
415
  "normalized": false,
416
  "rstrip": false,
417
  "single_word": false,
418
  "special": true
419
  },
420
- "36049": {
421
- "content": "<extra_id_46>",
422
  "lstrip": false,
423
  "normalized": false,
424
  "rstrip": false,
425
  "single_word": false,
426
  "special": true
427
  },
428
- "36050": {
429
- "content": "<extra_id_45>",
430
  "lstrip": false,
431
  "normalized": false,
432
  "rstrip": false,
433
  "single_word": false,
434
  "special": true
435
  },
436
- "36051": {
437
- "content": "<extra_id_44>",
438
  "lstrip": false,
439
  "normalized": false,
440
  "rstrip": false,
441
  "single_word": false,
442
  "special": true
443
  },
444
- "36052": {
445
- "content": "<extra_id_43>",
446
  "lstrip": false,
447
  "normalized": false,
448
  "rstrip": false,
449
  "single_word": false,
450
  "special": true
451
  },
452
- "36053": {
453
- "content": "<extra_id_42>",
454
  "lstrip": false,
455
  "normalized": false,
456
  "rstrip": false,
457
  "single_word": false,
458
  "special": true
459
  },
460
- "36054": {
461
- "content": "<extra_id_41>",
462
  "lstrip": false,
463
  "normalized": false,
464
  "rstrip": false,
465
  "single_word": false,
466
  "special": true
467
  },
468
- "36055": {
469
- "content": "<extra_id_40>",
470
  "lstrip": false,
471
  "normalized": false,
472
  "rstrip": false,
473
  "single_word": false,
474
  "special": true
475
  },
476
- "36056": {
477
- "content": "<extra_id_39>",
478
  "lstrip": false,
479
  "normalized": false,
480
  "rstrip": false,
481
  "single_word": false,
482
  "special": true
483
  },
484
- "36057": {
485
- "content": "<extra_id_38>",
486
  "lstrip": false,
487
  "normalized": false,
488
  "rstrip": false,
489
  "single_word": false,
490
  "special": true
491
  },
492
- "36058": {
493
- "content": "<extra_id_37>",
494
  "lstrip": false,
495
  "normalized": false,
496
  "rstrip": false,
497
  "single_word": false,
498
  "special": true
499
  },
500
- "36059": {
501
- "content": "<extra_id_36>",
502
  "lstrip": false,
503
  "normalized": false,
504
  "rstrip": false,
505
  "single_word": false,
506
  "special": true
507
  },
508
- "36060": {
509
- "content": "<extra_id_35>",
510
  "lstrip": false,
511
  "normalized": false,
512
  "rstrip": false,
513
  "single_word": false,
514
  "special": true
515
  },
516
- "36061": {
517
- "content": "<extra_id_34>",
518
  "lstrip": false,
519
  "normalized": false,
520
  "rstrip": false,
521
  "single_word": false,
522
  "special": true
523
  },
524
- "36062": {
525
- "content": "<extra_id_33>",
526
  "lstrip": false,
527
  "normalized": false,
528
  "rstrip": false,
529
  "single_word": false,
530
  "special": true
531
  },
532
- "36063": {
533
- "content": "<extra_id_32>",
534
  "lstrip": false,
535
  "normalized": false,
536
  "rstrip": false,
537
  "single_word": false,
538
  "special": true
539
  },
540
- "36064": {
541
- "content": "<extra_id_31>",
542
  "lstrip": false,
543
  "normalized": false,
544
  "rstrip": false,
545
  "single_word": false,
546
  "special": true
547
  },
548
- "36065": {
549
- "content": "<extra_id_30>",
550
  "lstrip": false,
551
  "normalized": false,
552
  "rstrip": false,
553
  "single_word": false,
554
  "special": true
555
  },
556
- "36066": {
557
- "content": "<extra_id_29>",
558
  "lstrip": false,
559
  "normalized": false,
560
  "rstrip": false,
561
  "single_word": false,
562
  "special": true
563
  },
564
- "36067": {
565
- "content": "<extra_id_28>",
566
  "lstrip": false,
567
  "normalized": false,
568
  "rstrip": false,
569
  "single_word": false,
570
  "special": true
571
  },
572
- "36068": {
573
- "content": "<extra_id_27>",
574
  "lstrip": false,
575
  "normalized": false,
576
  "rstrip": false,
577
  "single_word": false,
578
  "special": true
579
  },
580
- "36069": {
581
- "content": "<extra_id_26>",
582
  "lstrip": false,
583
  "normalized": false,
584
  "rstrip": false,
585
  "single_word": false,
586
  "special": true
587
  },
588
- "36070": {
589
- "content": "<extra_id_25>",
590
  "lstrip": false,
591
  "normalized": false,
592
  "rstrip": false,
593
  "single_word": false,
594
  "special": true
595
  },
596
- "36071": {
597
- "content": "<extra_id_24>",
598
  "lstrip": false,
599
  "normalized": false,
600
  "rstrip": false,
601
  "single_word": false,
602
  "special": true
603
  },
604
- "36072": {
605
- "content": "<extra_id_23>",
606
  "lstrip": false,
607
  "normalized": false,
608
  "rstrip": false,
609
  "single_word": false,
610
  "special": true
611
  },
612
- "36073": {
613
- "content": "<extra_id_22>",
614
  "lstrip": false,
615
  "normalized": false,
616
  "rstrip": false,
617
  "single_word": false,
618
  "special": true
619
  },
620
- "36074": {
621
- "content": "<extra_id_21>",
622
  "lstrip": false,
623
  "normalized": false,
624
  "rstrip": false,
625
  "single_word": false,
626
  "special": true
627
  },
628
- "36075": {
629
- "content": "<extra_id_20>",
630
  "lstrip": false,
631
  "normalized": false,
632
  "rstrip": false,
633
  "single_word": false,
634
  "special": true
635
  },
636
- "36076": {
637
- "content": "<extra_id_19>",
638
  "lstrip": false,
639
  "normalized": false,
640
  "rstrip": false,
641
  "single_word": false,
642
  "special": true
643
  },
644
- "36077": {
645
- "content": "<extra_id_18>",
646
  "lstrip": false,
647
  "normalized": false,
648
  "rstrip": false,
649
  "single_word": false,
650
  "special": true
651
  },
652
- "36078": {
653
- "content": "<extra_id_17>",
654
  "lstrip": false,
655
  "normalized": false,
656
  "rstrip": false,
657
  "single_word": false,
658
  "special": true
659
  },
660
- "36079": {
661
- "content": "<extra_id_16>",
662
  "lstrip": false,
663
  "normalized": false,
664
  "rstrip": false,
665
  "single_word": false,
666
  "special": true
667
  },
668
- "36080": {
669
- "content": "<extra_id_15>",
670
  "lstrip": false,
671
  "normalized": false,
672
  "rstrip": false,
673
  "single_word": false,
674
  "special": true
675
  },
676
- "36081": {
677
- "content": "<extra_id_14>",
678
  "lstrip": false,
679
  "normalized": false,
680
  "rstrip": false,
681
  "single_word": false,
682
  "special": true
683
  },
684
- "36082": {
685
- "content": "<extra_id_13>",
686
  "lstrip": false,
687
  "normalized": false,
688
  "rstrip": false,
689
  "single_word": false,
690
  "special": true
691
  },
692
- "36083": {
693
- "content": "<extra_id_12>",
694
  "lstrip": false,
695
  "normalized": false,
696
  "rstrip": false,
697
  "single_word": false,
698
  "special": true
699
  },
700
- "36084": {
701
- "content": "<extra_id_11>",
702
  "lstrip": false,
703
  "normalized": false,
704
  "rstrip": false,
705
  "single_word": false,
706
  "special": true
707
  },
708
- "36085": {
709
- "content": "<extra_id_10>",
710
  "lstrip": false,
711
  "normalized": false,
712
  "rstrip": false,
713
  "single_word": false,
714
  "special": true
715
  },
716
- "36086": {
717
- "content": "<extra_id_9>",
718
  "lstrip": false,
719
  "normalized": false,
720
  "rstrip": false,
721
  "single_word": false,
722
  "special": true
723
  },
724
- "36087": {
725
- "content": "<extra_id_8>",
726
  "lstrip": false,
727
  "normalized": false,
728
  "rstrip": false,
729
  "single_word": false,
730
  "special": true
731
  },
732
- "36088": {
733
- "content": "<extra_id_7>",
734
  "lstrip": false,
735
  "normalized": false,
736
  "rstrip": false,
737
  "single_word": false,
738
  "special": true
739
  },
740
- "36089": {
741
- "content": "<extra_id_6>",
742
  "lstrip": false,
743
  "normalized": false,
744
  "rstrip": false,
745
  "single_word": false,
746
  "special": true
747
  },
748
- "36090": {
749
- "content": "<extra_id_5>",
750
  "lstrip": false,
751
  "normalized": false,
752
  "rstrip": false,
753
  "single_word": false,
754
  "special": true
755
  },
756
- "36091": {
757
- "content": "<extra_id_4>",
758
  "lstrip": false,
759
  "normalized": false,
760
  "rstrip": false,
761
  "single_word": false,
762
  "special": true
763
  },
764
- "36092": {
765
- "content": "<extra_id_3>",
766
  "lstrip": false,
767
  "normalized": false,
768
  "rstrip": false,
769
  "single_word": false,
770
  "special": true
771
  },
772
- "36093": {
773
- "content": "<extra_id_2>",
774
  "lstrip": false,
775
  "normalized": false,
776
  "rstrip": false,
777
  "single_word": false,
778
  "special": true
779
  },
780
- "36094": {
781
- "content": "<extra_id_1>",
782
  "lstrip": false,
783
  "normalized": false,
784
  "rstrip": false,
785
  "single_word": false,
786
  "special": true
787
  },
788
- "36095": {
789
- "content": "<extra_id_0>",
790
- "lstrip": false,
791
- "normalized": false,
792
  "rstrip": false,
793
  "single_word": false,
794
  "special": true
795
  }
796
  },
797
  "additional_special_tokens": [
798
- "<extra_id_0>",
799
- "<extra_id_1>",
800
- "<extra_id_2>",
801
- "<extra_id_3>",
802
- "<extra_id_4>",
803
- "<extra_id_5>",
804
- "<extra_id_6>",
805
- "<extra_id_7>",
806
- "<extra_id_8>",
807
- "<extra_id_9>",
808
- "<extra_id_10>",
809
- "<extra_id_11>",
810
- "<extra_id_12>",
811
- "<extra_id_13>",
812
- "<extra_id_14>",
813
- "<extra_id_15>",
814
- "<extra_id_16>",
815
- "<extra_id_17>",
816
- "<extra_id_18>",
817
- "<extra_id_19>",
818
- "<extra_id_20>",
819
- "<extra_id_21>",
820
- "<extra_id_22>",
821
- "<extra_id_23>",
822
- "<extra_id_24>",
823
- "<extra_id_25>",
824
- "<extra_id_26>",
825
- "<extra_id_27>",
826
- "<extra_id_28>",
827
- "<extra_id_29>",
828
- "<extra_id_30>",
829
- "<extra_id_31>",
830
- "<extra_id_32>",
831
- "<extra_id_33>",
832
- "<extra_id_34>",
833
- "<extra_id_35>",
834
- "<extra_id_36>",
835
- "<extra_id_37>",
836
- "<extra_id_38>",
837
- "<extra_id_39>",
838
- "<extra_id_40>",
839
- "<extra_id_41>",
840
- "<extra_id_42>",
841
- "<extra_id_43>",
842
- "<extra_id_44>",
843
- "<extra_id_45>",
844
- "<extra_id_46>",
845
- "<extra_id_47>",
846
- "<extra_id_48>",
847
- "<extra_id_49>",
848
- "<extra_id_50>",
849
- "<extra_id_51>",
850
- "<extra_id_52>",
851
- "<extra_id_53>",
852
- "<extra_id_54>",
853
- "<extra_id_55>",
854
- "<extra_id_56>",
855
- "<extra_id_57>",
856
- "<extra_id_58>",
857
- "<extra_id_59>",
858
- "<extra_id_60>",
859
- "<extra_id_61>",
860
- "<extra_id_62>",
861
- "<extra_id_63>",
862
- "<extra_id_64>",
863
- "<extra_id_65>",
864
- "<extra_id_66>",
865
- "<extra_id_67>",
866
- "<extra_id_68>",
867
- "<extra_id_69>",
868
- "<extra_id_70>",
869
- "<extra_id_71>",
870
- "<extra_id_72>",
871
- "<extra_id_73>",
872
- "<extra_id_74>",
873
- "<extra_id_75>",
874
- "<extra_id_76>",
875
- "<extra_id_77>",
876
- "<extra_id_78>",
877
- "<extra_id_79>",
878
- "<extra_id_80>",
879
- "<extra_id_81>",
880
- "<extra_id_82>",
881
- "<extra_id_83>",
882
- "<extra_id_84>",
883
- "<extra_id_85>",
884
- "<extra_id_86>",
885
- "<extra_id_87>",
886
- "<extra_id_88>",
887
- "<extra_id_89>",
888
- "<extra_id_90>",
889
- "<extra_id_91>",
890
- "<extra_id_92>",
891
- "<extra_id_93>",
892
- "<extra_id_94>",
893
- "<extra_id_95>"
894
  ],
 
895
  "clean_up_tokenization_spaces": false,
 
896
  "eos_token": "</s>",
897
- "extra_ids": 96,
898
  "extra_special_tokens": {},
899
- "model_max_length": 1000000000000000019884624838656,
 
 
 
900
  "pad_token": "<pad>",
 
 
 
901
  "sp_model_kwargs": {},
902
- "tokenizer_class": "T5Tokenizer",
 
 
 
 
 
903
  "unk_token": "<unk>"
904
  }
 
1
  {
 
2
  "added_tokens_decoder": {
3
  "0": {
4
+ "content": "<s>",
5
  "lstrip": false,
6
  "normalized": false,
7
  "rstrip": false,
 
9
  "special": true
10
  },
11
  "1": {
12
+ "content": "<pad>",
13
  "lstrip": false,
14
  "normalized": false,
15
  "rstrip": false,
 
17
  "special": true
18
  },
19
  "2": {
20
+ "content": "</s>",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21
  "lstrip": false,
22
  "normalized": false,
23
  "rstrip": false,
24
  "single_word": false,
25
  "special": true
26
  },
27
+ "3": {
28
+ "content": "<unk>",
29
  "lstrip": false,
30
  "normalized": false,
31
  "rstrip": false,
32
  "single_word": false,
33
  "special": true
34
  },
35
+ "250001": {
36
+ "content": "ar_AR",
37
  "lstrip": false,
38
  "normalized": false,
39
  "rstrip": false,
40
  "single_word": false,
41
  "special": true
42
  },
43
+ "250002": {
44
+ "content": "cs_CZ",
45
  "lstrip": false,
46
  "normalized": false,
47
  "rstrip": false,
48
  "single_word": false,
49
  "special": true
50
  },
51
+ "250003": {
52
+ "content": "de_DE",
53
  "lstrip": false,
54
  "normalized": false,
55
  "rstrip": false,
56
  "single_word": false,
57
  "special": true
58
  },
59
+ "250004": {
60
+ "content": "en_XX",
61
  "lstrip": false,
62
  "normalized": false,
63
  "rstrip": false,
64
  "single_word": false,
65
  "special": true
66
  },
67
+ "250005": {
68
+ "content": "es_XX",
69
  "lstrip": false,
70
  "normalized": false,
71
  "rstrip": false,
72
  "single_word": false,
73
  "special": true
74
  },
75
+ "250006": {
76
+ "content": "et_EE",
77
  "lstrip": false,
78
  "normalized": false,
79
  "rstrip": false,
80
  "single_word": false,
81
  "special": true
82
  },
83
+ "250007": {
84
+ "content": "fi_FI",
85
  "lstrip": false,
86
  "normalized": false,
87
  "rstrip": false,
88
  "single_word": false,
89
  "special": true
90
  },
91
+ "250008": {
92
+ "content": "fr_XX",
93
  "lstrip": false,
94
  "normalized": false,
95
  "rstrip": false,
96
  "single_word": false,
97
  "special": true
98
  },
99
+ "250009": {
100
+ "content": "gu_IN",
101
  "lstrip": false,
102
  "normalized": false,
103
  "rstrip": false,
104
  "single_word": false,
105
  "special": true
106
  },
107
+ "250010": {
108
+ "content": "hi_IN",
109
  "lstrip": false,
110
  "normalized": false,
111
  "rstrip": false,
112
  "single_word": false,
113
  "special": true
114
  },
115
+ "250011": {
116
+ "content": "it_IT",
117
  "lstrip": false,
118
  "normalized": false,
119
  "rstrip": false,
120
  "single_word": false,
121
  "special": true
122
  },
123
+ "250012": {
124
+ "content": "ja_XX",
125
  "lstrip": false,
126
  "normalized": false,
127
  "rstrip": false,
128
  "single_word": false,
129
  "special": true
130
  },
131
+ "250013": {
132
+ "content": "kk_KZ",
133
  "lstrip": false,
134
  "normalized": false,
135
  "rstrip": false,
136
  "single_word": false,
137
  "special": true
138
  },
139
+ "250014": {
140
+ "content": "ko_KR",
141
  "lstrip": false,
142
  "normalized": false,
143
  "rstrip": false,
144
  "single_word": false,
145
  "special": true
146
  },
147
+ "250015": {
148
+ "content": "lt_LT",
149
  "lstrip": false,
150
  "normalized": false,
151
  "rstrip": false,
152
  "single_word": false,
153
  "special": true
154
  },
155
+ "250016": {
156
+ "content": "lv_LV",
157
  "lstrip": false,
158
  "normalized": false,
159
  "rstrip": false,
160
  "single_word": false,
161
  "special": true
162
  },
163
+ "250017": {
164
+ "content": "my_MM",
165
  "lstrip": false,
166
  "normalized": false,
167
  "rstrip": false,
168
  "single_word": false,
169
  "special": true
170
  },
171
+ "250018": {
172
+ "content": "ne_NP",
173
  "lstrip": false,
174
  "normalized": false,
175
  "rstrip": false,
176
  "single_word": false,
177
  "special": true
178
  },
179
+ "250019": {
180
+ "content": "nl_XX",
181
  "lstrip": false,
182
  "normalized": false,
183
  "rstrip": false,
184
  "single_word": false,
185
  "special": true
186
  },
187
+ "250020": {
188
+ "content": "ro_RO",
189
  "lstrip": false,
190
  "normalized": false,
191
  "rstrip": false,
192
  "single_word": false,
193
  "special": true
194
  },
195
+ "250021": {
196
+ "content": "ru_RU",
197
  "lstrip": false,
198
  "normalized": false,
199
  "rstrip": false,
200
  "single_word": false,
201
  "special": true
202
  },
203
+ "250022": {
204
+ "content": "si_LK",
205
  "lstrip": false,
206
  "normalized": false,
207
  "rstrip": false,
208
  "single_word": false,
209
  "special": true
210
  },
211
+ "250023": {
212
+ "content": "tr_TR",
213
  "lstrip": false,
214
  "normalized": false,
215
  "rstrip": false,
216
  "single_word": false,
217
  "special": true
218
  },
219
+ "250024": {
220
+ "content": "vi_VN",
221
  "lstrip": false,
222
  "normalized": false,
223
  "rstrip": false,
224
  "single_word": false,
225
  "special": true
226
  },
227
+ "250025": {
228
+ "content": "zh_CN",
229
  "lstrip": false,
230
  "normalized": false,
231
  "rstrip": false,
232
  "single_word": false,
233
  "special": true
234
  },
235
+ "250026": {
236
+ "content": "af_ZA",
237
  "lstrip": false,
238
  "normalized": false,
239
  "rstrip": false,
240
  "single_word": false,
241
  "special": true
242
  },
243
+ "250027": {
244
+ "content": "az_AZ",
245
  "lstrip": false,
246
  "normalized": false,
247
  "rstrip": false,
248
  "single_word": false,
249
  "special": true
250
  },
251
+ "250028": {
252
+ "content": "bn_IN",
253
  "lstrip": false,
254
  "normalized": false,
255
  "rstrip": false,
256
  "single_word": false,
257
  "special": true
258
  },
259
+ "250029": {
260
+ "content": "fa_IR",
261
  "lstrip": false,
262
  "normalized": false,
263
  "rstrip": false,
264
  "single_word": false,
265
  "special": true
266
  },
267
+ "250030": {
268
+ "content": "he_IL",
269
  "lstrip": false,
270
  "normalized": false,
271
  "rstrip": false,
272
  "single_word": false,
273
  "special": true
274
  },
275
+ "250031": {
276
+ "content": "hr_HR",
277
  "lstrip": false,
278
  "normalized": false,
279
  "rstrip": false,
280
  "single_word": false,
281
  "special": true
282
  },
283
+ "250032": {
284
+ "content": "id_ID",
285
  "lstrip": false,
286
  "normalized": false,
287
  "rstrip": false,
288
  "single_word": false,
289
  "special": true
290
  },
291
+ "250033": {
292
+ "content": "ka_GE",
293
  "lstrip": false,
294
  "normalized": false,
295
  "rstrip": false,
296
  "single_word": false,
297
  "special": true
298
  },
299
+ "250034": {
300
+ "content": "km_KH",
301
  "lstrip": false,
302
  "normalized": false,
303
  "rstrip": false,
304
  "single_word": false,
305
  "special": true
306
  },
307
+ "250035": {
308
+ "content": "mk_MK",
309
  "lstrip": false,
310
  "normalized": false,
311
  "rstrip": false,
312
  "single_word": false,
313
  "special": true
314
  },
315
+ "250036": {
316
+ "content": "ml_IN",
317
  "lstrip": false,
318
  "normalized": false,
319
  "rstrip": false,
320
  "single_word": false,
321
  "special": true
322
  },
323
+ "250037": {
324
+ "content": "mn_MN",
325
  "lstrip": false,
326
  "normalized": false,
327
  "rstrip": false,
328
  "single_word": false,
329
  "special": true
330
  },
331
+ "250038": {
332
+ "content": "mr_IN",
333
  "lstrip": false,
334
  "normalized": false,
335
  "rstrip": false,
336
  "single_word": false,
337
  "special": true
338
  },
339
+ "250039": {
340
+ "content": "pl_PL",
341
  "lstrip": false,
342
  "normalized": false,
343
  "rstrip": false,
344
  "single_word": false,
345
  "special": true
346
  },
347
+ "250040": {
348
+ "content": "ps_AF",
349
  "lstrip": false,
350
  "normalized": false,
351
  "rstrip": false,
352
  "single_word": false,
353
  "special": true
354
  },
355
+ "250041": {
356
+ "content": "pt_XX",
357
  "lstrip": false,
358
  "normalized": false,
359
  "rstrip": false,
360
  "single_word": false,
361
  "special": true
362
  },
363
+ "250042": {
364
+ "content": "sv_SE",
365
  "lstrip": false,
366
  "normalized": false,
367
  "rstrip": false,
368
  "single_word": false,
369
  "special": true
370
  },
371
+ "250043": {
372
+ "content": "sw_KE",
373
  "lstrip": false,
374
  "normalized": false,
375
  "rstrip": false,
376
  "single_word": false,
377
  "special": true
378
  },
379
+ "250044": {
380
+ "content": "ta_IN",
381
  "lstrip": false,
382
  "normalized": false,
383
  "rstrip": false,
384
  "single_word": false,
385
  "special": true
386
  },
387
+ "250045": {
388
+ "content": "te_IN",
389
  "lstrip": false,
390
  "normalized": false,
391
  "rstrip": false,
392
  "single_word": false,
393
  "special": true
394
  },
395
+ "250046": {
396
+ "content": "th_TH",
397
  "lstrip": false,
398
  "normalized": false,
399
  "rstrip": false,
400
  "single_word": false,
401
  "special": true
402
  },
403
+ "250047": {
404
+ "content": "tl_XX",
405
  "lstrip": false,
406
  "normalized": false,
407
  "rstrip": false,
408
  "single_word": false,
409
  "special": true
410
  },
411
+ "250048": {
412
+ "content": "uk_UA",
413
  "lstrip": false,
414
  "normalized": false,
415
  "rstrip": false,
416
  "single_word": false,
417
  "special": true
418
  },
419
+ "250049": {
420
+ "content": "ur_PK",
421
  "lstrip": false,
422
  "normalized": false,
423
  "rstrip": false,
424
  "single_word": false,
425
  "special": true
426
  },
427
+ "250050": {
428
+ "content": "xh_ZA",
429
  "lstrip": false,
430
  "normalized": false,
431
  "rstrip": false,
432
  "single_word": false,
433
  "special": true
434
  },
435
+ "250051": {
436
+ "content": "gl_ES",
437
  "lstrip": false,
438
  "normalized": false,
439
  "rstrip": false,
440
  "single_word": false,
441
  "special": true
442
  },
443
+ "250052": {
444
+ "content": "sl_SI",
445
  "lstrip": false,
446
  "normalized": false,
447
  "rstrip": false,
448
  "single_word": false,
449
  "special": true
450
  },
451
+ "250053": {
452
+ "content": "<mask>",
453
+ "lstrip": true,
454
+ "normalized": true,
455
  "rstrip": false,
456
  "single_word": false,
457
  "special": true
458
  }
459
  },
460
  "additional_special_tokens": [
461
+ "ar_AR",
462
+ "cs_CZ",
463
+ "de_DE",
464
+ "en_XX",
465
+ "es_XX",
466
+ "et_EE",
467
+ "fi_FI",
468
+ "fr_XX",
469
+ "gu_IN",
470
+ "hi_IN",
471
+ "it_IT",
472
+ "ja_XX",
473
+ "kk_KZ",
474
+ "ko_KR",
475
+ "lt_LT",
476
+ "lv_LV",
477
+ "my_MM",
478
+ "ne_NP",
479
+ "nl_XX",
480
+ "ro_RO",
481
+ "ru_RU",
482
+ "si_LK",
483
+ "tr_TR",
484
+ "vi_VN",
485
+ "zh_CN",
486
+ "af_ZA",
487
+ "az_AZ",
488
+ "bn_IN",
489
+ "fa_IR",
490
+ "he_IL",
491
+ "hr_HR",
492
+ "id_ID",
493
+ "ka_GE",
494
+ "km_KH",
495
+ "mk_MK",
496
+ "ml_IN",
497
+ "mn_MN",
498
+ "mr_IN",
499
+ "pl_PL",
500
+ "ps_AF",
501
+ "pt_XX",
502
+ "sv_SE",
503
+ "sw_KE",
504
+ "ta_IN",
505
+ "te_IN",
506
+ "th_TH",
507
+ "tl_XX",
508
+ "uk_UA",
509
+ "ur_PK",
510
+ "xh_ZA",
511
+ "gl_ES",
512
+ "sl_SI"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
513
  ],
514
+ "bos_token": "<s>",
515
  "clean_up_tokenization_spaces": false,
516
+ "cls_token": "<s>",
517
  "eos_token": "</s>",
 
518
  "extra_special_tokens": {},
519
+ "mask_token": "<mask>",
520
+ "max_length": 50,
521
+ "model_max_length": 1024,
522
+ "pad_to_multiple_of": null,
523
  "pad_token": "<pad>",
524
+ "pad_token_type_id": 0,
525
+ "padding_side": "right",
526
+ "sep_token": "</s>",
527
  "sp_model_kwargs": {},
528
+ "src_lang": "vi_VN",
529
+ "stride": 0,
530
+ "tgt_lang": "vi_VN",
531
+ "tokenizer_class": "MBart50Tokenizer",
532
+ "truncation_side": "right",
533
+ "truncation_strategy": "longest_first",
534
  "unk_token": "<unk>"
535
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0e1ade7795c4e7edaf77613fc9d06dfa2f2549b92d0c77338ecdacd349c4f3ca
3
  size 5969
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43619790f43bee12f7f55f0a682f1e8d9588ef07e77f1f5690c53febc38c5f31
3
  size 5969