Hugging Face's logo

guymorlan
/
TokenizerLabeller

Text2Text Generation
Transformers
PyTorch
marian
Inference Endpoints
Model card Files Community
1
guymorlan commited on
Commit
ad8ab3a
·
1 Parent(s): ad31383

add updated model files, new architecture

Browse files
Files changed (6) hide show
  1. playaling_words.json +0 -0
  2. pytorch_model.bin +1 -1
  3. rng_state.pth +3 -0
  4. scheduler.pt +3 -0
  5. trainer_state.json +213 -311
  6. training_args.bin +1 -1
playaling_words.json DELETED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7e16a00f3b7f5e5a385a1ce482a06886aaa158f7810c2d751cc6c06faba66b9d
3
  size 308321413
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6a46a29427f2b7642e604221ccc3a99c7344d84b80ecb503770cc76eb6db8c8
3
  size 308321413
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a8af7acfeb1e1d6a4eb4294cdf97bc63445757ad0d0aa78a19b5d357906885c
3
+ size 14575
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52ee137ddf3e7c84e65c0bbe4aa2de396127a8cd6c5276dbbba60ed19c1bcc8f
3
+ size 627
trainer_state.json CHANGED
@@ -1,377 +1,279 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 24.983073798239676,
5
- "global_step": 3075,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 1.0,
12
- "eval_accuracy": 0.3861538956236535,
13
- "eval_bleu": 58.02299639521676,
14
- "eval_loss": 0.3069784641265869,
15
- "eval_perfect_accuracy": 0.0773067331670823,
16
- "eval_runtime": 1188.6804,
17
- "eval_samples_per_second": 3.036,
18
- "eval_simple_accuracy": 0.6000514485996334,
19
- "eval_steps_per_second": 0.051,
20
- "eval_token_accuracy": 0.4887938518923438,
21
- "step": 123
22
  },
23
  {
24
  "epoch": 2.0,
25
- "eval_accuracy": 0.4548114329807414,
26
- "eval_bleu": 65.34000395485323,
27
- "eval_loss": 0.21667715907096863,
28
- "eval_perfect_accuracy": 0.10446106954835134,
29
- "eval_runtime": 1040.2638,
30
- "eval_samples_per_second": 3.469,
31
- "eval_simple_accuracy": 0.6507732373082983,
32
- "eval_steps_per_second": 0.059,
33
- "eval_token_accuracy": 0.5523261421727808,
34
- "step": 246
35
  },
36
  {
37
- "epoch": 3.0,
38
- "eval_accuracy": 0.49269371274754853,
39
- "eval_bleu": 68.74113419920226,
40
- "eval_loss": 0.18049955368041992,
41
- "eval_perfect_accuracy": 0.11831532280410086,
42
- "eval_runtime": 1030.6108,
43
- "eval_samples_per_second": 3.502,
44
- "eval_simple_accuracy": 0.667243478818176,
45
- "eval_steps_per_second": 0.059,
46
- "eval_token_accuracy": 0.5896942895597,
47
- "step": 369
48
- },
49
- {
50
- "epoch": 4.0,
51
- "eval_accuracy": 0.5115833811985449,
52
- "eval_bleu": 70.81806278651923,
53
- "eval_loss": 0.1621144413948059,
54
- "eval_perfect_accuracy": 0.13022998060404545,
55
- "eval_runtime": 1015.4894,
56
- "eval_samples_per_second": 3.554,
57
- "eval_simple_accuracy": 0.6820154445082647,
58
- "eval_steps_per_second": 0.06,
59
- "eval_token_accuracy": 0.6063884102367733,
60
- "step": 492
61
- },
62
- {
63
- "epoch": 4.06,
64
- "learning_rate": 4.186991869918699e-05,
65
- "loss": 0.6023,
66
  "step": 500
67
  },
68
  {
69
- "epoch": 5.0,
70
- "eval_accuracy": 0.528468272733082,
71
- "eval_bleu": 72.98982499035274,
72
- "eval_loss": 0.15042348206043243,
73
- "eval_perfect_accuracy": 0.13577168190634525,
74
- "eval_runtime": 1033.5998,
75
- "eval_samples_per_second": 3.492,
76
- "eval_simple_accuracy": 0.6923445587577481,
77
- "eval_steps_per_second": 0.059,
78
- "eval_token_accuracy": 0.6224039874752381,
79
- "step": 615
80
- },
81
- {
82
- "epoch": 6.0,
83
- "eval_accuracy": 0.5492471073070218,
84
- "eval_bleu": 74.56394107515801,
85
- "eval_loss": 0.14072531461715698,
86
- "eval_perfect_accuracy": 0.14657799944582986,
87
- "eval_runtime": 1023.0499,
88
- "eval_samples_per_second": 3.528,
89
- "eval_simple_accuracy": 0.7019495958155016,
90
- "eval_steps_per_second": 0.06,
91
- "eval_token_accuracy": 0.6440957362498019,
92
- "step": 738
93
- },
94
- {
95
- "epoch": 7.0,
96
- "eval_accuracy": 0.564568482688391,
97
- "eval_bleu": 75.61656636338704,
98
- "eval_loss": 0.13451269268989563,
99
- "eval_perfect_accuracy": 0.14934885009697976,
100
- "eval_runtime": 1024.191,
101
- "eval_samples_per_second": 3.524,
102
- "eval_simple_accuracy": 0.7093622708757638,
103
- "eval_steps_per_second": 0.06,
104
- "eval_token_accuracy": 0.6577774949083504,
105
- "step": 861
106
  },
107
  {
108
- "epoch": 7.99,
109
- "eval_accuracy": 0.5669532973180321,
110
- "eval_bleu": 76.49084761884065,
111
- "eval_loss": 0.13158030807971954,
112
- "eval_perfect_accuracy": 0.14768633970628983,
113
- "eval_runtime": 1031.2577,
114
- "eval_samples_per_second": 3.5,
115
- "eval_simple_accuracy": 0.7093948790077678,
116
- "eval_steps_per_second": 0.059,
117
- "eval_token_accuracy": 0.661253716075824,
118
- "step": 984
119
  },
120
  {
121
- "epoch": 8.12,
122
- "learning_rate": 3.373983739837399e-05,
123
- "loss": 0.1587,
124
  "step": 1000
125
  },
126
  {
127
- "epoch": 8.99,
128
- "eval_accuracy": 0.5799008390541571,
129
- "eval_bleu": 77.12937097778037,
130
- "eval_loss": 0.12549354135990143,
131
- "eval_perfect_accuracy": 0.16098642283180936,
132
- "eval_runtime": 1023.3584,
133
- "eval_samples_per_second": 3.527,
134
- "eval_simple_accuracy": 0.7213640986524281,
135
- "eval_steps_per_second": 0.06,
136
- "eval_token_accuracy": 0.6679379608441394,
137
- "step": 1107
138
- },
139
- {
140
- "epoch": 9.99,
141
- "eval_accuracy": 0.5873817235028894,
142
- "eval_bleu": 77.65888247465986,
143
- "eval_loss": 0.12297692894935608,
144
- "eval_perfect_accuracy": 0.16181767802715433,
145
- "eval_runtime": 1023.9043,
146
- "eval_samples_per_second": 3.525,
147
- "eval_simple_accuracy": 0.7230901124023623,
148
- "eval_steps_per_second": 0.06,
149
- "eval_token_accuracy": 0.6762557947545564,
150
- "step": 1230
151
  },
152
  {
153
- "epoch": 10.99,
154
- "eval_accuracy": 0.59429296131896,
155
- "eval_bleu": 78.01343118892193,
156
- "eval_loss": 0.12066468596458435,
157
- "eval_perfect_accuracy": 0.171515655306179,
158
- "eval_runtime": 1021.8584,
159
- "eval_samples_per_second": 3.532,
160
- "eval_simple_accuracy": 0.7276157260621433,
161
- "eval_steps_per_second": 0.06,
162
- "eval_token_accuracy": 0.6818008877615727,
163
  "step": 1353
164
  },
165
  {
166
- "epoch": 12.0,
167
- "eval_accuracy": 0.6051917739086551,
168
- "eval_bleu": 78.9493434619066,
169
- "eval_loss": 0.11863834410905838,
170
- "eval_perfect_accuracy": 0.17594901634801885,
171
- "eval_runtime": 1010.8004,
172
- "eval_samples_per_second": 3.57,
173
- "eval_simple_accuracy": 0.7328412818571789,
174
- "eval_steps_per_second": 0.06,
175
- "eval_token_accuracy": 0.6936664143325764,
176
- "step": 1477
177
- },
178
- {
179
- "epoch": 12.19,
180
- "learning_rate": 2.5609756097560977e-05,
181
- "loss": 0.1271,
182
  "step": 1500
183
  },
184
  {
185
- "epoch": 13.0,
186
- "eval_accuracy": 0.6046047631112237,
187
- "eval_bleu": 79.73687831704848,
188
- "eval_loss": 0.11646737158298492,
189
- "eval_perfect_accuracy": 0.17650318647824884,
190
- "eval_runtime": 1016.1798,
191
- "eval_samples_per_second": 3.552,
192
- "eval_simple_accuracy": 0.733183430453509,
193
- "eval_steps_per_second": 0.06,
194
- "eval_token_accuracy": 0.6896060298961236,
195
- "step": 1600
196
- },
197
- {
198
- "epoch": 14.0,
199
- "eval_accuracy": 0.6090549311381985,
200
- "eval_bleu": 79.73048335421603,
201
- "eval_loss": 0.11454274505376816,
202
- "eval_perfect_accuracy": 0.17955112219451372,
203
- "eval_runtime": 1021.0729,
204
- "eval_samples_per_second": 3.535,
205
- "eval_simple_accuracy": 0.7366154820326104,
206
- "eval_steps_per_second": 0.06,
207
- "eval_token_accuracy": 0.6950767769510844,
208
- "step": 1723
209
  },
210
  {
211
- "epoch": 15.0,
212
- "eval_accuracy": 0.6126954921803128,
213
- "eval_bleu": 79.95232854185171,
214
- "eval_loss": 0.11355158686637878,
215
- "eval_perfect_accuracy": 0.18149071765031866,
216
- "eval_runtime": 1019.1512,
217
- "eval_samples_per_second": 3.541,
218
- "eval_simple_accuracy": 0.738064270532627,
219
- "eval_steps_per_second": 0.06,
220
- "eval_token_accuracy": 0.6964121435142594,
221
- "step": 1846
222
  },
223
  {
224
- "epoch": 16.0,
225
- "eval_accuracy": 0.6142770873724692,
226
- "eval_bleu": 80.17800049091872,
227
- "eval_loss": 0.11310549825429916,
228
- "eval_perfect_accuracy": 0.17871986699916875,
229
- "eval_runtime": 1000.0367,
230
- "eval_samples_per_second": 3.609,
231
- "eval_simple_accuracy": 0.7383911260846073,
232
- "eval_steps_per_second": 0.061,
233
- "eval_token_accuracy": 0.6993293710072148,
234
- "step": 1969
235
  },
236
  {
237
- "epoch": 16.25,
238
- "learning_rate": 1.747967479674797e-05,
239
- "loss": 0.1109,
240
- "step": 2000
 
 
 
 
 
 
 
241
  },
242
  {
243
- "epoch": 17.0,
244
- "eval_accuracy": 0.6165277159040286,
245
- "eval_bleu": 80.11868616750657,
246
- "eval_loss": 0.11200028657913208,
247
- "eval_perfect_accuracy": 0.17899695206428373,
248
- "eval_runtime": 994.6715,
249
- "eval_samples_per_second": 3.628,
250
- "eval_simple_accuracy": 0.7413924775663463,
251
- "eval_steps_per_second": 0.061,
252
- "eval_token_accuracy": 0.6986571628587793,
253
- "step": 2092
254
  },
255
  {
256
- "epoch": 18.0,
257
- "eval_accuracy": 0.622873435767464,
258
- "eval_bleu": 80.72391323382952,
259
- "eval_loss": 0.1101086214184761,
260
- "eval_perfect_accuracy": 0.18703241895261846,
261
- "eval_runtime": 1022.8027,
262
- "eval_samples_per_second": 3.529,
263
- "eval_simple_accuracy": 0.7436717883731981,
264
- "eval_steps_per_second": 0.06,
265
- "eval_token_accuracy": 0.7055916363060352,
266
- "step": 2215
267
  },
268
  {
269
- "epoch": 19.0,
270
- "eval_accuracy": 0.6222941661905669,
271
- "eval_bleu": 80.76914319903946,
272
- "eval_loss": 0.11085215210914612,
273
- "eval_perfect_accuracy": 0.18647824882238848,
274
- "eval_runtime": 1021.7653,
275
- "eval_samples_per_second": 3.532,
276
- "eval_simple_accuracy": 0.7428426331492414,
277
- "eval_steps_per_second": 0.06,
278
- "eval_token_accuracy": 0.7042150701453691,
279
- "step": 2338
280
  },
281
  {
282
- "epoch": 19.99,
283
- "eval_accuracy": 0.6245798719005644,
284
- "eval_bleu": 80.78924704054619,
285
- "eval_loss": 0.11003228276968002,
286
- "eval_perfect_accuracy": 0.19229703518980326,
287
- "eval_runtime": 1016.0765,
288
- "eval_samples_per_second": 3.552,
289
- "eval_simple_accuracy": 0.7465914135328809,
290
- "eval_steps_per_second": 0.06,
291
- "eval_token_accuracy": 0.7053395903354683,
292
- "step": 2461
293
  },
294
  {
295
- "epoch": 20.31,
296
- "learning_rate": 9.34959349593496e-06,
297
- "loss": 0.1009,
298
- "step": 2500
 
 
 
 
 
 
 
299
  },
300
  {
301
- "epoch": 20.99,
302
- "eval_accuracy": 0.6269267364414843,
303
- "eval_bleu": 81.12519074317272,
304
- "eval_loss": 0.10916706919670105,
305
- "eval_perfect_accuracy": 0.1886949293433084,
306
- "eval_runtime": 1038.5849,
307
- "eval_samples_per_second": 3.475,
308
- "eval_simple_accuracy": 0.7457342213764668,
309
- "eval_steps_per_second": 0.059,
310
- "eval_token_accuracy": 0.708499841420869,
311
- "step": 2584
312
  },
313
  {
314
- "epoch": 21.99,
315
- "eval_accuracy": 0.6295486935866983,
316
- "eval_bleu": 81.28558003169574,
317
- "eval_loss": 0.10952065885066986,
318
- "eval_perfect_accuracy": 0.1886949293433084,
319
- "eval_runtime": 1037.9043,
320
- "eval_samples_per_second": 3.477,
321
- "eval_simple_accuracy": 0.7463499604117181,
322
- "eval_steps_per_second": 0.059,
323
- "eval_token_accuracy": 0.7121773555027712,
324
- "step": 2707
325
  },
326
  {
327
- "epoch": 22.99,
328
- "eval_accuracy": 0.6287138141188421,
329
- "eval_bleu": 80.93511268093461,
330
- "eval_loss": 0.1084115132689476,
331
- "eval_perfect_accuracy": 0.19479080077583819,
332
- "eval_runtime": 1032.8176,
333
- "eval_samples_per_second": 3.494,
334
- "eval_simple_accuracy": 0.7478732859319451,
335
- "eval_steps_per_second": 0.059,
336
- "eval_token_accuracy": 0.7086084306754697,
337
- "step": 2830
338
  },
339
  {
340
- "epoch": 24.0,
341
- "eval_accuracy": 0.6307677671011321,
342
- "eval_bleu": 81.11966959311552,
343
- "eval_loss": 0.10846742242574692,
344
- "eval_perfect_accuracy": 0.19201995012468828,
345
- "eval_runtime": 1015.3974,
346
- "eval_samples_per_second": 3.554,
347
- "eval_simple_accuracy": 0.7478197443947611,
348
- "eval_steps_per_second": 0.06,
349
- "eval_token_accuracy": 0.7126502394317065,
350
- "step": 2954
351
  },
352
  {
353
- "epoch": 24.37,
354
- "learning_rate": 1.2195121951219514e-06,
355
- "loss": 0.0953,
356
- "step": 3000
 
 
 
 
 
 
 
357
  },
358
  {
359
- "epoch": 24.98,
360
- "eval_accuracy": 0.6323030591218893,
361
- "eval_bleu": 81.29208359004284,
362
- "eval_loss": 0.10844007134437561,
363
- "eval_perfect_accuracy": 0.19257412025491827,
364
- "eval_runtime": 1029.0079,
365
- "eval_samples_per_second": 3.507,
366
- "eval_simple_accuracy": 0.7477571722935489,
367
- "eval_steps_per_second": 0.059,
368
- "eval_token_accuracy": 0.7139641781581867,
369
- "step": 3075
370
  }
371
  ],
372
- "max_steps": 3075,
373
  "num_train_epochs": 25,
374
- "total_flos": 9686690034941952.0,
375
  "trial_name": null,
376
  "trial_params": null
377
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 16.995936461026968,
5
+ "global_step": 3834,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 1.0,
12
+ "eval_bleu": 69.52679870560905,
13
+ "eval_loss": 0.16526192426681519,
14
+ "eval_perfect_accuracy": 0.10141313383208644,
15
+ "eval_runtime": 3340.3895,
16
+ "eval_samples_per_second": 1.08,
17
+ "eval_steps_per_second": 0.045,
18
+ "eval_subword_accuracy": 0.5594287933042841,
19
+ "eval_token_accuracy": 0.6964062297769533,
20
+ "eval_word_accuracy": 0.4744973429025366,
21
+ "step": 225
22
  },
23
  {
24
  "epoch": 2.0,
25
+ "eval_bleu": 74.31204559162617,
26
+ "eval_loss": 0.11631891876459122,
27
+ "eval_perfect_accuracy": 0.14990302022720975,
28
+ "eval_runtime": 3269.4249,
29
+ "eval_samples_per_second": 1.104,
30
+ "eval_steps_per_second": 0.046,
31
+ "eval_subword_accuracy": 0.6142826689405244,
32
+ "eval_token_accuracy": 0.7389895544659987,
33
+ "eval_word_accuracy": 0.532967032967033,
34
+ "step": 451
35
  },
36
  {
37
+ "epoch": 2.22,
38
+ "learning_rate": 4.555555555555556e-05,
39
+ "loss": 0.6045,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
40
  "step": 500
41
  },
42
  {
43
+ "epoch": 3.0,
44
+ "eval_bleu": 76.85411902074411,
45
+ "eval_loss": 0.09909386187791824,
46
+ "eval_perfect_accuracy": 0.18952618453865336,
47
+ "eval_runtime": 3381.7443,
48
+ "eval_samples_per_second": 1.067,
49
+ "eval_steps_per_second": 0.045,
50
+ "eval_subword_accuracy": 0.6512524901453821,
51
+ "eval_token_accuracy": 0.7695926757936676,
52
+ "eval_word_accuracy": 0.5709718317623064,
53
+ "step": 676
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
54
  },
55
  {
56
+ "epoch": 4.0,
57
+ "eval_bleu": 79.14575222404268,
58
+ "eval_loss": 0.09123498946428299,
59
+ "eval_perfect_accuracy": 0.21169298974785258,
60
+ "eval_runtime": 3480.4117,
61
+ "eval_samples_per_second": 1.037,
62
+ "eval_steps_per_second": 0.043,
63
+ "eval_subword_accuracy": 0.6773669070746654,
64
+ "eval_token_accuracy": 0.7950973443948806,
65
+ "eval_word_accuracy": 0.5997376762487704,
66
+ "step": 902
67
  },
68
  {
69
+ "epoch": 4.43,
70
+ "learning_rate": 4.111111111111111e-05,
71
+ "loss": 0.1137,
72
  "step": 1000
73
  },
74
  {
75
+ "epoch": 5.0,
76
+ "eval_bleu": 80.22128752790911,
77
+ "eval_loss": 0.08361164480447769,
78
+ "eval_perfect_accuracy": 0.23330562482682182,
79
+ "eval_runtime": 1727.4258,
80
+ "eval_samples_per_second": 2.089,
81
+ "eval_steps_per_second": 0.087,
82
+ "eval_subword_accuracy": 0.693643785404865,
83
+ "eval_token_accuracy": 0.8034821726091302,
84
+ "eval_word_accuracy": 0.6188809276376962,
85
+ "step": 1127
 
 
 
 
 
 
 
 
 
 
 
 
 
86
  },
87
  {
88
+ "epoch": 6.0,
89
+ "eval_bleu": 80.47331007551738,
90
+ "eval_loss": 0.07888099551200867,
91
+ "eval_perfect_accuracy": 0.25270157938487114,
92
+ "eval_runtime": 1706.4952,
93
+ "eval_samples_per_second": 2.115,
94
+ "eval_steps_per_second": 0.088,
95
+ "eval_subword_accuracy": 0.6985128247973933,
96
+ "eval_token_accuracy": 0.8036176790040939,
97
+ "eval_word_accuracy": 0.6238234160841085,
98
  "step": 1353
99
  },
100
  {
101
+ "epoch": 6.65,
102
+ "learning_rate": 3.6666666666666666e-05,
103
+ "loss": 0.0886,
 
 
 
 
 
 
 
 
 
 
 
 
 
104
  "step": 1500
105
  },
106
  {
107
+ "epoch": 7.0,
108
+ "eval_bleu": 81.35496460228585,
109
+ "eval_loss": 0.07820562273263931,
110
+ "eval_perfect_accuracy": 0.257689110556941,
111
+ "eval_runtime": 1696.8822,
112
+ "eval_samples_per_second": 2.127,
113
+ "eval_steps_per_second": 0.089,
114
+ "eval_subword_accuracy": 0.7077631303620004,
115
+ "eval_token_accuracy": 0.8164888051893702,
116
+ "eval_word_accuracy": 0.6347371318469585,
117
+ "step": 1579
 
 
 
 
 
 
 
 
 
 
 
 
 
118
  },
119
  {
120
+ "epoch": 8.0,
121
+ "eval_bleu": 81.4155970043395,
122
+ "eval_loss": 0.07889819890260696,
123
+ "eval_perfect_accuracy": 0.2454973676918814,
124
+ "eval_runtime": 1666.9192,
125
+ "eval_samples_per_second": 2.165,
126
+ "eval_steps_per_second": 0.091,
127
+ "eval_subword_accuracy": 0.6984033613445378,
128
+ "eval_token_accuracy": 0.8176890756302521,
129
+ "eval_word_accuracy": 0.6222492535662943,
130
+ "step": 1804
131
  },
132
  {
133
+ "epoch": 8.87,
134
+ "learning_rate": 3.222222222222223e-05,
135
+ "loss": 0.0768,
136
+ "step": 2000
 
 
 
 
 
 
 
137
  },
138
  {
139
+ "epoch": 9.0,
140
+ "eval_bleu": 81.70774187369526,
141
+ "eval_loss": 0.07279801368713379,
142
+ "eval_perfect_accuracy": 0.27847049044056527,
143
+ "eval_runtime": 1589.4085,
144
+ "eval_samples_per_second": 2.271,
145
+ "eval_steps_per_second": 0.095,
146
+ "eval_subword_accuracy": 0.7123310810810811,
147
+ "eval_token_accuracy": 0.8106841216216216,
148
+ "eval_word_accuracy": 0.6412536119137586,
149
+ "step": 2030
150
  },
151
  {
152
+ "epoch": 10.0,
153
+ "eval_bleu": 82.60753055516308,
154
+ "eval_loss": 0.07139766961336136,
155
+ "eval_perfect_accuracy": 0.29897478525907456,
156
+ "eval_runtime": 1630.9624,
157
+ "eval_samples_per_second": 2.213,
158
+ "eval_steps_per_second": 0.093,
159
+ "eval_subword_accuracy": 0.7325649377766642,
160
+ "eval_token_accuracy": 0.8289902280130294,
161
+ "eval_word_accuracy": 0.6648324437352116,
162
+ "step": 2255
163
  },
164
  {
165
+ "epoch": 11.0,
166
+ "eval_bleu": 83.62166331521699,
167
+ "eval_loss": 0.06930351257324219,
168
+ "eval_perfect_accuracy": 0.30950401773344416,
169
+ "eval_runtime": 1612.5857,
170
+ "eval_samples_per_second": 2.238,
171
+ "eval_steps_per_second": 0.094,
172
+ "eval_subword_accuracy": 0.739740238184157,
173
+ "eval_token_accuracy": 0.8358438109465123,
174
+ "eval_word_accuracy": 0.6716549295774648,
175
+ "step": 2481
176
  },
177
  {
178
+ "epoch": 11.08,
179
+ "learning_rate": 2.777777777777778e-05,
180
+ "loss": 0.0681,
181
+ "step": 2500
 
 
 
 
 
 
 
182
  },
183
  {
184
+ "epoch": 12.0,
185
+ "eval_bleu": 84.02807089662639,
186
+ "eval_loss": 0.06798376142978668,
187
+ "eval_perfect_accuracy": 0.3081185924078692,
188
+ "eval_runtime": 1680.6877,
189
+ "eval_samples_per_second": 2.147,
190
+ "eval_steps_per_second": 0.09,
191
+ "eval_subword_accuracy": 0.745770423991727,
192
+ "eval_token_accuracy": 0.8428541882109617,
193
+ "eval_word_accuracy": 0.6793930762489044,
194
+ "step": 2707
195
  },
196
  {
197
+ "epoch": 13.0,
198
+ "eval_bleu": 84.42354785113844,
199
+ "eval_loss": 0.06762586534023285,
200
+ "eval_perfect_accuracy": 0.31698531449154893,
201
+ "eval_runtime": 1751.9665,
202
+ "eval_samples_per_second": 2.06,
203
+ "eval_steps_per_second": 0.086,
204
+ "eval_subword_accuracy": 0.7447953313190679,
205
+ "eval_token_accuracy": 0.8407350689127105,
206
+ "eval_word_accuracy": 0.67777227994942,
207
+ "step": 2932
208
  },
209
  {
210
+ "epoch": 13.3,
211
+ "learning_rate": 2.3333333333333336e-05,
212
+ "loss": 0.0619,
213
+ "step": 3000
 
 
 
 
 
 
 
214
  },
215
  {
216
+ "epoch": 14.0,
217
+ "eval_bleu": 84.59233008431931,
218
+ "eval_loss": 0.06648800522089005,
219
+ "eval_perfect_accuracy": 0.32696037683568857,
220
+ "eval_runtime": 1738.746,
221
+ "eval_samples_per_second": 2.076,
222
+ "eval_steps_per_second": 0.087,
223
+ "eval_subword_accuracy": 0.7536297454036935,
224
+ "eval_token_accuracy": 0.8476946489532349,
225
+ "eval_word_accuracy": 0.6883173734610123,
226
+ "step": 3158
227
  },
228
  {
229
+ "epoch": 15.0,
230
+ "eval_bleu": 84.49240456076029,
231
+ "eval_loss": 0.06621185690164566,
232
+ "eval_perfect_accuracy": 0.32917705735660846,
233
+ "eval_runtime": 1720.559,
234
+ "eval_samples_per_second": 2.098,
235
+ "eval_steps_per_second": 0.088,
236
+ "eval_subword_accuracy": 0.7496996063807748,
237
+ "eval_token_accuracy": 0.8439196188108556,
238
+ "eval_word_accuracy": 0.6839415256100242,
239
+ "step": 3383
240
  },
241
  {
242
+ "epoch": 15.52,
243
+ "learning_rate": 1.888888888888889e-05,
244
+ "loss": 0.0574,
245
+ "step": 3500
 
 
 
 
 
 
 
246
  },
247
  {
248
+ "epoch": 16.0,
249
+ "eval_bleu": 85.02924908908608,
250
+ "eval_loss": 0.06474177539348602,
251
+ "eval_perfect_accuracy": 0.343585480742588,
252
+ "eval_runtime": 1678.6132,
253
+ "eval_samples_per_second": 2.15,
254
+ "eval_steps_per_second": 0.09,
255
+ "eval_subword_accuracy": 0.7591370348981107,
256
+ "eval_token_accuracy": 0.8501773781041169,
257
+ "eval_word_accuracy": 0.6949701058636388,
258
+ "step": 3609
259
  },
260
  {
261
+ "epoch": 17.0,
262
+ "eval_bleu": 84.79034637013908,
263
+ "eval_loss": 0.06463250517845154,
264
+ "eval_perfect_accuracy": 0.34968135217511775,
265
+ "eval_runtime": 1711.9697,
266
+ "eval_samples_per_second": 2.108,
267
+ "eval_steps_per_second": 0.088,
268
+ "eval_subword_accuracy": 0.762099607356892,
269
+ "eval_token_accuracy": 0.8493490390576566,
270
+ "eval_word_accuracy": 0.7003675462175655,
271
+ "step": 3834
272
  }
273
  ],
274
+ "max_steps": 5625,
275
  "num_train_epochs": 25,
276
+ "total_flos": 6098403774431232.0,
277
  "trial_name": null,
278
  "trial_params": null
279
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9fb8a43a695feecbdd75975e3545d24ccd9aed989a14888afb3e50c07eee1c23
3
  size 4219
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ade5cc7246ed0457ed40a82cf7baea9db562870c60bef51dc078fc52843a21b
3
  size 4219