Hugging Face's logo

sn4kebyt3
/
ru-bart-large

Text2Text Generation
Transformers
PyTorch
Safetensors
Russian
English
mbart
mbart-50
Inference Endpoints
Model card Files Community
1
sn4kebyt3 commited on
Commit
321ef25
•
1 Parent(s): 298d910

Upload 6 files

Browse files
Files changed (5) hide show
  1. config.json +2 -2
  2. generation_config.json +1 -1
  3. pytorch_model.bin +2 -2
  4. special_tokens_map.json +51 -1
  5. tokenizer_config.json +52 -2
config.json CHANGED
@@ -52,7 +52,7 @@
52
  "static_position_embeddings": false,
53
  "tokenizer_class": "MBart50Tokenizer",
54
  "torch_dtype": "float32",
55
- "transformers_version": "4.27.4",
56
  "use_cache": true,
57
- "vocab_size": 24211
58
  }
 
52
  "static_position_embeddings": false,
53
  "tokenizer_class": "MBart50Tokenizer",
54
  "torch_dtype": "float32",
55
+ "transformers_version": "4.28.1",
56
  "use_cache": true,
57
+ "vocab_size": 24261
58
  }
generation_config.json CHANGED
@@ -8,5 +8,5 @@
8
  "max_length": 200,
9
  "num_beams": 5,
10
  "pad_token_id": 1,
11
- "transformers_version": "4.27.4"
12
  }
 
8
  "max_length": 200,
9
  "num_beams": 5,
10
  "pad_token_id": 1,
11
+ "transformers_version": "4.28.1"
12
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dc02b923217a856397d121e8eb373314baa6797326503914da8f51105197c690
3
- size 1518730461
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24287dd4538893ba3e23fa51d7f2b960550885a52969815baf5c4666a05dcac0
3
+ size 1518935453
special_tokens_map.json CHANGED
@@ -1,7 +1,57 @@
1
  {
2
  "additional_special_tokens": [
 
 
 
3
  "en_XX",
4
- "ru_RU"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5
  ],
6
  "bos_token": "<s>",
7
  "cls_token": "<s>",
 
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": "<s>",
57
  "cls_token": "<s>",
tokenizer_config.json CHANGED
@@ -1,9 +1,60 @@
1
  {
2
  "additional_special_tokens": [
 
 
 
3
  "en_XX",
4
- "ru_RU"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5
  ],
6
  "bos_token": "<s>",
 
7
  "cls_token": "<s>",
8
  "eos_token": "</s>",
9
  "mask_token": {
@@ -18,7 +69,6 @@
18
  "pad_token": "<pad>",
19
  "sep_token": "</s>",
20
  "sp_model_kwargs": {},
21
- "special_tokens_map_file": "/home/suraj/projects/mbart-50/hf_models/mbart-50-large/special_tokens_map.json",
22
  "src_lang": null,
23
  "tgt_lang": null,
24
  "tokenizer_class": "MBart50Tokenizer",
 
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": "<s>",
57
+ "clean_up_tokenization_spaces": true,
58
  "cls_token": "<s>",
59
  "eos_token": "</s>",
60
  "mask_token": {
 
69
  "pad_token": "<pad>",
70
  "sep_token": "</s>",
71
  "sp_model_kwargs": {},
 
72
  "src_lang": null,
73
  "tgt_lang": null,
74
  "tokenizer_class": "MBart50Tokenizer",