Hugging Face's logo

Everlyn
/
transformer_en_de10

Transformers
PyTorch
mbart
text2text-generation
Model card Files
xet
 Community
1
Everlyn commited on
Commit
d95a1fe
·
1 Parent(s): c3c16ee

model

Browse files
Files changed (11) hide show
  1. .gitattributes +1 -0
  2. config.json +36 -0
  3. eval_results.json +10 -0
  4. generation_config.json +8 -0
  5. pytorch_model.bin +3 -0
  6. sentencepiece.bpe.model +3 -0
  7. special_tokens_map.json +63 -0
  8. tokenizer.json +3 -0
  9. tokenizer_config.json +76 -0
  10. train_results.json +8 -0
  11. trainer_state.json +217 -0
.gitattributes CHANGED
@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
35
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Everlyn/transformer_base",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "relu",
5
+ "architectures": [
6
+ "MBartForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "bos_token_id": 0,
10
+ "classifier_dropout": 0.0,
11
+ "d_model": 512,
12
+ "decoder_attention_heads": 8,
13
+ "decoder_ffn_dim": 2048,
14
+ "decoder_layerdrop": 0.0,
15
+ "decoder_layers": 6,
16
+ "decoder_start_token_id": 250003,
17
+ "dropout": 0.1,
18
+ "encoder_attention_heads": 8,
19
+ "encoder_ffn_dim": 2048,
20
+ "encoder_layerdrop": 0.0,
21
+ "encoder_layers": 6,
22
+ "eos_token_id": 2,
23
+ "forced_bos_token_id": 250003,
24
+ "forced_eos_token_id": 2,
25
+ "init_std": 0.02,
26
+ "is_encoder_decoder": true,
27
+ "max_position_embeddings": 512,
28
+ "model_type": "mbart",
29
+ "num_hidden_layers": 6,
30
+ "pad_token_id": 1,
31
+ "scale_embedding": false,
32
+ "torch_dtype": "float32",
33
+ "transformers_version": "4.27.2",
34
+ "use_cache": true,
35
+ "vocab_size": 250054
36
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "eval_bleu": 12.6624,
4
+ "eval_gen_len": 22.0831,
5
+ "eval_loss": 2.0509557723999023,
6
+ "eval_runtime": 206.2459,
7
+ "eval_samples": 2998,
8
+ "eval_samples_per_second": 14.536,
9
+ "eval_steps_per_second": 0.184
10
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 2,
5
+ "forced_eos_token_id": 2,
6
+ "pad_token_id": 1,
7
+ "transformers_version": "4.27.2"
8
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33b3048a2268e32ef715f1b2baaf99d2816e70173d795819132bf00bd067e989
3
+ size 691874225
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 ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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>",
58
+ "eos_token": "</s>",
59
+ "mask_token": "<mask>",
60
+ "pad_token": "<pad>",
61
+ "sep_token": "</s>",
62
+ "unk_token": "<unk>"
63
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d58a68c276b56fcc48c165c63f70e5e4d452b4182032a5f7a2d018f4aa1a889
3
+ size 17109752
tokenizer_config.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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>",
58
+ "eos_token": "</s>",
59
+ "mask_token": {
60
+ "__type": "AddedToken",
61
+ "content": "<mask>",
62
+ "lstrip": true,
63
+ "normalized": true,
64
+ "rstrip": false,
65
+ "single_word": false
66
+ },
67
+ "model_max_length": 1024,
68
+ "pad_token": "<pad>",
69
+ "sep_token": "</s>",
70
+ "sp_model_kwargs": {},
71
+ "special_tokens_map_file": "/home/suraj/projects/mbart-50/hf_models/mbart-50-large/special_tokens_map.json",
72
+ "src_lang": null,
73
+ "tgt_lang": null,
74
+ "tokenizer_class": "MBart50Tokenizer",
75
+ "unk_token": "<unk>"
76
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "train_loss": 2.836493848511954,
4
+ "train_runtime": 50326.1237,
5
+ "train_samples": 3869033,
6
+ "train_samples_per_second": 307.517,
7
+ "train_steps_per_second": 0.32
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,217 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.9997518764344644,
5
+ "global_step": 16120,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.12,
12
+ "learning_rate": 4.8449131513647645e-05,
13
+ "loss": 7.5778,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.25,
18
+ "learning_rate": 4.689826302729529e-05,
19
+ "loss": 5.6514,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.37,
24
+ "learning_rate": 4.534739454094293e-05,
25
+ "loss": 4.753,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.5,
30
+ "learning_rate": 4.379652605459057e-05,
31
+ "loss": 4.1203,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.62,
36
+ "learning_rate": 4.2245657568238216e-05,
37
+ "loss": 3.6995,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.74,
42
+ "learning_rate": 4.069478908188586e-05,
43
+ "loss": 3.4027,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.87,
48
+ "learning_rate": 3.91439205955335e-05,
49
+ "loss": 3.1636,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.99,
54
+ "learning_rate": 3.7593052109181144e-05,
55
+ "loss": 2.9862,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 1.12,
60
+ "learning_rate": 3.6042183622828787e-05,
61
+ "loss": 2.8469,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 1.24,
66
+ "learning_rate": 3.449131513647643e-05,
67
+ "loss": 2.7431,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 1.36,
72
+ "learning_rate": 3.294044665012407e-05,
73
+ "loss": 2.6552,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 1.49,
78
+ "learning_rate": 3.1389578163771715e-05,
79
+ "loss": 2.5856,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 1.61,
84
+ "learning_rate": 2.9838709677419357e-05,
85
+ "loss": 2.5195,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 1.74,
90
+ "learning_rate": 2.8287841191067e-05,
91
+ "loss": 2.4654,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 1.86,
96
+ "learning_rate": 2.6736972704714643e-05,
97
+ "loss": 2.4238,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 1.98,
102
+ "learning_rate": 2.5186104218362282e-05,
103
+ "loss": 2.3821,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 2.11,
108
+ "learning_rate": 2.3635235732009928e-05,
109
+ "loss": 2.3322,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 2.23,
114
+ "learning_rate": 2.208436724565757e-05,
115
+ "loss": 2.3004,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 2.36,
120
+ "learning_rate": 2.053349875930521e-05,
121
+ "loss": 2.2745,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 2.48,
126
+ "learning_rate": 1.8982630272952853e-05,
127
+ "loss": 2.2494,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 2.61,
132
+ "learning_rate": 1.74317617866005e-05,
133
+ "loss": 2.2309,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 2.73,
138
+ "learning_rate": 1.588089330024814e-05,
139
+ "loss": 2.2102,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 2.85,
144
+ "learning_rate": 1.4330024813895782e-05,
145
+ "loss": 2.1912,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 2.98,
150
+ "learning_rate": 1.2779156327543427e-05,
151
+ "loss": 2.1743,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 3.1,
156
+ "learning_rate": 1.1228287841191068e-05,
157
+ "loss": 2.1509,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 3.23,
162
+ "learning_rate": 9.67741935483871e-06,
163
+ "loss": 2.1385,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 3.35,
168
+ "learning_rate": 8.126550868486353e-06,
169
+ "loss": 2.1343,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 3.47,
174
+ "learning_rate": 6.575682382133995e-06,
175
+ "loss": 2.1247,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 3.6,
180
+ "learning_rate": 5.0248138957816385e-06,
181
+ "loss": 2.1188,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 3.72,
186
+ "learning_rate": 3.4739454094292807e-06,
187
+ "loss": 2.1198,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 3.85,
192
+ "learning_rate": 1.9230769230769234e-06,
193
+ "loss": 2.1102,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 3.97,
198
+ "learning_rate": 3.7220843672456576e-07,
199
+ "loss": 2.106,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 4.0,
204
+ "step": 16120,
205
+ "total_flos": 4.540075318875832e+17,
206
+ "train_loss": 2.836493848511954,
207
+ "train_runtime": 50326.1237,
208
+ "train_samples_per_second": 307.517,
209
+ "train_steps_per_second": 0.32
210
+ }
211
+ ],
212
+ "max_steps": 16120,
213
+ "num_train_epochs": 4,
214
+ "total_flos": 4.540075318875832e+17,
215
+ "trial_name": null,
216
+ "trial_params": null
217
+ }