AhmedSSoliman commited on
Commit
0e61c45
·
1 Parent(s): 96d3bdf

Upload trainer_state.json

Browse files
Files changed (1) hide show
  1. trainer_state.json +358 -0
trainer_state.json ADDED
@@ -0,0 +1,358 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.06889855116605759,
3
+ "best_model_checkpoint": "./models/1/checkpoint-9737",
4
+ "epoch": 10.0,
5
+ "global_step": 13910,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.36,
12
+ "learning_rate": 3.591954022988506e-05,
13
+ "loss": 0.0876,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.72,
18
+ "learning_rate": 4.884970485848343e-05,
19
+ "loss": 0.0879,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 1.0,
24
+ "eval_BLEU": 8.064725756490624,
25
+ "eval_BLEU-Bigram-Precision": 10.432874977383753,
26
+ "eval_BLEU-Trigram-Precision": 5.407291933269737,
27
+ "eval_BLEU-Unigram-Precision": 21.631135458605595,
28
+ "eval_ROUGE-2": 12.39926519031995,
29
+ "eval_ROUGE-L": 24.731962013373813,
30
+ "eval_Sacre-Bigram-Precision": 8.552870639534884,
31
+ "eval_Sacre-Trigram-Precision": 4.851033999298983,
32
+ "eval_Sacre-Unigram-Precision": 19.295765314011796,
33
+ "eval_SacreBLEU": 7.1292786955103695,
34
+ "eval_loss": 0.0844602882862091,
35
+ "eval_runtime": 523.4808,
36
+ "eval_samples_per_second": 2.363,
37
+ "eval_steps_per_second": 0.296,
38
+ "step": 1391
39
+ },
40
+ {
41
+ "epoch": 1.08,
42
+ "learning_rate": 4.6961555925533526e-05,
43
+ "loss": 0.0829,
44
+ "step": 1500
45
+ },
46
+ {
47
+ "epoch": 1.44,
48
+ "learning_rate": 4.506962312698653e-05,
49
+ "loss": 0.0747,
50
+ "step": 2000
51
+ },
52
+ {
53
+ "epoch": 1.8,
54
+ "learning_rate": 4.317769032843954e-05,
55
+ "loss": 0.0738,
56
+ "step": 2500
57
+ },
58
+ {
59
+ "epoch": 2.0,
60
+ "eval_BLEU": 7.726645595861129,
61
+ "eval_BLEU-Bigram-Precision": 9.779983147645352,
62
+ "eval_BLEU-Trigram-Precision": 5.376859377396105,
63
+ "eval_BLEU-Unigram-Precision": 19.00918707221551,
64
+ "eval_ROUGE-2": 14.567403936228773,
65
+ "eval_ROUGE-L": 26.962145889007623,
66
+ "eval_Sacre-Bigram-Precision": 8.340425531914894,
67
+ "eval_Sacre-Trigram-Precision": 5.06410008133545,
68
+ "eval_Sacre-Unigram-Precision": 17.33626552335896,
69
+ "eval_SacreBLEU": 7.08253441311445,
70
+ "eval_loss": 0.07733357697725296,
71
+ "eval_runtime": 594.316,
72
+ "eval_samples_per_second": 2.081,
73
+ "eval_steps_per_second": 0.261,
74
+ "step": 2782
75
+ },
76
+ {
77
+ "epoch": 2.16,
78
+ "learning_rate": 4.1285757529892535e-05,
79
+ "loss": 0.0669,
80
+ "step": 3000
81
+ },
82
+ {
83
+ "epoch": 2.52,
84
+ "learning_rate": 3.939382473134555e-05,
85
+ "loss": 0.0631,
86
+ "step": 3500
87
+ },
88
+ {
89
+ "epoch": 2.88,
90
+ "learning_rate": 3.7505675798395645e-05,
91
+ "loss": 0.0616,
92
+ "step": 4000
93
+ },
94
+ {
95
+ "epoch": 3.0,
96
+ "eval_BLEU": 13.75944614514078,
97
+ "eval_BLEU-Bigram-Precision": 17.05563909774436,
98
+ "eval_BLEU-Trigram-Precision": 9.895979758223222,
99
+ "eval_BLEU-Unigram-Precision": 31.226259170122074,
100
+ "eval_ROUGE-2": 16.762154637582928,
101
+ "eval_ROUGE-L": 29.733231785857146,
102
+ "eval_Sacre-Bigram-Precision": 14.784847370356749,
103
+ "eval_Sacre-Trigram-Precision": 9.445382434455736,
104
+ "eval_Sacre-Unigram-Precision": 28.888232688616565,
105
+ "eval_SacreBLEU": 12.80566461172721,
106
+ "eval_loss": 0.07288952171802521,
107
+ "eval_runtime": 656.3315,
108
+ "eval_samples_per_second": 1.885,
109
+ "eval_steps_per_second": 0.236,
110
+ "step": 4173
111
+ },
112
+ {
113
+ "epoch": 3.24,
114
+ "learning_rate": 3.561374299984865e-05,
115
+ "loss": 0.0556,
116
+ "step": 4500
117
+ },
118
+ {
119
+ "epoch": 3.59,
120
+ "learning_rate": 3.372181020130165e-05,
121
+ "loss": 0.0522,
122
+ "step": 5000
123
+ },
124
+ {
125
+ "epoch": 3.95,
126
+ "learning_rate": 3.1829877402754655e-05,
127
+ "loss": 0.0541,
128
+ "step": 5500
129
+ },
130
+ {
131
+ "epoch": 4.0,
132
+ "eval_BLEU": 12.584990506585802,
133
+ "eval_BLEU-Bigram-Precision": 15.284490474363894,
134
+ "eval_BLEU-Trigram-Precision": 9.189817259955635,
135
+ "eval_BLEU-Unigram-Precision": 27.802290416935204,
136
+ "eval_ROUGE-2": 18.2427503756888,
137
+ "eval_ROUGE-L": 31.248445460832464,
138
+ "eval_Sacre-Bigram-Precision": 13.375350140056023,
139
+ "eval_Sacre-Trigram-Precision": 8.759613065730592,
140
+ "eval_Sacre-Unigram-Precision": 25.743221169552434,
141
+ "eval_SacreBLEU": 11.731554660360798,
142
+ "eval_loss": 0.07118678092956543,
143
+ "eval_runtime": 463.4231,
144
+ "eval_samples_per_second": 2.669,
145
+ "eval_steps_per_second": 0.334,
146
+ "step": 5564
147
+ },
148
+ {
149
+ "epoch": 4.31,
150
+ "learning_rate": 2.9937944604207657e-05,
151
+ "loss": 0.0464,
152
+ "step": 6000
153
+ },
154
+ {
155
+ "epoch": 4.67,
156
+ "learning_rate": 2.8046011805660665e-05,
157
+ "loss": 0.0459,
158
+ "step": 6500
159
+ },
160
+ {
161
+ "epoch": 5.0,
162
+ "eval_BLEU": 15.1776190124192,
163
+ "eval_BLEU-Bigram-Precision": 18.32776890695311,
164
+ "eval_BLEU-Trigram-Precision": 11.247289809684414,
165
+ "eval_BLEU-Unigram-Precision": 32.42811501597444,
166
+ "eval_ROUGE-2": 17.936785005872903,
167
+ "eval_ROUGE-L": 31.307371141735135,
168
+ "eval_Sacre-Bigram-Precision": 16.26334519572954,
169
+ "eval_Sacre-Trigram-Precision": 10.753317119724164,
170
+ "eval_Sacre-Unigram-Precision": 30.265755070515205,
171
+ "eval_SacreBLEU": 14.242434580808037,
172
+ "eval_loss": 0.06986960768699646,
173
+ "eval_runtime": 398.2357,
174
+ "eval_samples_per_second": 3.106,
175
+ "eval_steps_per_second": 0.389,
176
+ "step": 6955
177
+ },
178
+ {
179
+ "epoch": 5.03,
180
+ "learning_rate": 2.6154079007113667e-05,
181
+ "loss": 0.0455,
182
+ "step": 7000
183
+ },
184
+ {
185
+ "epoch": 5.39,
186
+ "learning_rate": 2.4262146208566673e-05,
187
+ "loss": 0.0393,
188
+ "step": 7500
189
+ },
190
+ {
191
+ "epoch": 5.75,
192
+ "learning_rate": 2.2370213410019678e-05,
193
+ "loss": 0.0405,
194
+ "step": 8000
195
+ },
196
+ {
197
+ "epoch": 6.0,
198
+ "eval_BLEU": 17.03804435280673,
199
+ "eval_BLEU-Bigram-Precision": 20.16791440608614,
200
+ "eval_BLEU-Trigram-Precision": 12.76924045849603,
201
+ "eval_BLEU-Unigram-Precision": 34.79696172947707,
202
+ "eval_ROUGE-2": 20.00404092356461,
203
+ "eval_ROUGE-L": 33.44410880996549,
204
+ "eval_Sacre-Bigram-Precision": 18.140892297184017,
205
+ "eval_Sacre-Trigram-Precision": 12.388609066253391,
206
+ "eval_Sacre-Unigram-Precision": 32.66758356754171,
207
+ "eval_SacreBLEU": 16.183245622226913,
208
+ "eval_loss": 0.06919773668050766,
209
+ "eval_runtime": 389.7744,
210
+ "eval_samples_per_second": 3.174,
211
+ "eval_steps_per_second": 0.398,
212
+ "step": 8346
213
+ },
214
+ {
215
+ "epoch": 6.11,
216
+ "learning_rate": 2.047828061147268e-05,
217
+ "loss": 0.0395,
218
+ "step": 8500
219
+ },
220
+ {
221
+ "epoch": 6.47,
222
+ "learning_rate": 1.8586347812925686e-05,
223
+ "loss": 0.0351,
224
+ "step": 9000
225
+ },
226
+ {
227
+ "epoch": 6.83,
228
+ "learning_rate": 1.669441501437869e-05,
229
+ "loss": 0.0356,
230
+ "step": 9500
231
+ },
232
+ {
233
+ "epoch": 7.0,
234
+ "eval_BLEU": 18.45635959837559,
235
+ "eval_BLEU-Bigram-Precision": 21.599746072052056,
236
+ "eval_BLEU-Trigram-Precision": 13.98176291793313,
237
+ "eval_BLEU-Unigram-Precision": 36.946429662207564,
238
+ "eval_ROUGE-2": 20.663766964343406,
239
+ "eval_ROUGE-L": 33.85196384000605,
240
+ "eval_Sacre-Bigram-Precision": 19.43390720747009,
241
+ "eval_Sacre-Trigram-Precision": 13.581706410862664,
242
+ "eval_Sacre-Unigram-Precision": 34.678927680798004,
243
+ "eval_SacreBLEU": 17.534804534318486,
244
+ "eval_loss": 0.06889855116605759,
245
+ "eval_runtime": 289.9351,
246
+ "eval_samples_per_second": 4.266,
247
+ "eval_steps_per_second": 0.535,
248
+ "step": 9737
249
+ },
250
+ {
251
+ "epoch": 7.19,
252
+ "learning_rate": 1.4802482215831695e-05,
253
+ "loss": 0.0324,
254
+ "step": 10000
255
+ },
256
+ {
257
+ "epoch": 7.55,
258
+ "learning_rate": 1.2910549417284698e-05,
259
+ "loss": 0.032,
260
+ "step": 10500
261
+ },
262
+ {
263
+ "epoch": 7.91,
264
+ "learning_rate": 1.1018616618737702e-05,
265
+ "loss": 0.0316,
266
+ "step": 11000
267
+ },
268
+ {
269
+ "epoch": 8.0,
270
+ "eval_BLEU": 23.036610051092996,
271
+ "eval_BLEU-Bigram-Precision": 26.881516587677723,
272
+ "eval_BLEU-Trigram-Precision": 17.706261436868488,
273
+ "eval_BLEU-Unigram-Precision": 44.332174416920175,
274
+ "eval_ROUGE-2": 21.29088469404566,
275
+ "eval_ROUGE-L": 35.036834725440904,
276
+ "eval_Sacre-Bigram-Precision": 24.360966513868387,
277
+ "eval_Sacre-Trigram-Precision": 17.249540910018364,
278
+ "eval_Sacre-Unigram-Precision": 41.78435079135624,
279
+ "eval_SacreBLEU": 21.988824762940034,
280
+ "eval_loss": 0.0689924880862236,
281
+ "eval_runtime": 259.2001,
282
+ "eval_samples_per_second": 4.772,
283
+ "eval_steps_per_second": 0.598,
284
+ "step": 11128
285
+ },
286
+ {
287
+ "epoch": 8.27,
288
+ "learning_rate": 9.126683820190707e-06,
289
+ "loss": 0.0295,
290
+ "step": 11500
291
+ },
292
+ {
293
+ "epoch": 8.63,
294
+ "learning_rate": 7.238534887240806e-06,
295
+ "loss": 0.0283,
296
+ "step": 12000
297
+ },
298
+ {
299
+ "epoch": 8.99,
300
+ "learning_rate": 5.3466020886938095e-06,
301
+ "loss": 0.0287,
302
+ "step": 12500
303
+ },
304
+ {
305
+ "epoch": 9.0,
306
+ "eval_BLEU": 26.995788994612766,
307
+ "eval_BLEU-Bigram-Precision": 30.950083340643914,
308
+ "eval_BLEU-Trigram-Precision": 20.98232920550995,
309
+ "eval_BLEU-Unigram-Precision": 50.522155190347405,
310
+ "eval_ROUGE-2": 22.105956241069418,
311
+ "eval_ROUGE-L": 35.85099346344104,
312
+ "eval_Sacre-Bigram-Precision": 28.175123119324088,
313
+ "eval_Sacre-Trigram-Precision": 20.573315467075037,
314
+ "eval_Sacre-Unigram-Precision": 47.89045785194694,
315
+ "eval_SacreBLEU": 25.924953054404842,
316
+ "eval_loss": 0.06922509521245956,
317
+ "eval_runtime": 198.6398,
318
+ "eval_samples_per_second": 6.227,
319
+ "eval_steps_per_second": 0.78,
320
+ "step": 12519
321
+ },
322
+ {
323
+ "epoch": 9.35,
324
+ "learning_rate": 3.454669290146814e-06,
325
+ "loss": 0.0271,
326
+ "step": 13000
327
+ },
328
+ {
329
+ "epoch": 9.71,
330
+ "learning_rate": 1.5627364915998186e-06,
331
+ "loss": 0.0265,
332
+ "step": 13500
333
+ },
334
+ {
335
+ "epoch": 10.0,
336
+ "eval_BLEU": 26.712396231914138,
337
+ "eval_BLEU-Bigram-Precision": 30.61851211072664,
338
+ "eval_BLEU-Trigram-Precision": 20.719279806242287,
339
+ "eval_BLEU-Unigram-Precision": 49.895307303855155,
340
+ "eval_ROUGE-2": 22.458649124626863,
341
+ "eval_ROUGE-L": 36.256812107261126,
342
+ "eval_Sacre-Bigram-Precision": 27.962698554345884,
343
+ "eval_Sacre-Trigram-Precision": 20.387249114521843,
344
+ "eval_Sacre-Unigram-Precision": 47.333925324538036,
345
+ "eval_SacreBLEU": 25.70732578879903,
346
+ "eval_loss": 0.06933248043060303,
347
+ "eval_runtime": 190.4819,
348
+ "eval_samples_per_second": 6.494,
349
+ "eval_steps_per_second": 0.814,
350
+ "step": 13910
351
+ }
352
+ ],
353
+ "max_steps": 13910,
354
+ "num_train_epochs": 10,
355
+ "total_flos": 3.016954478592e+16,
356
+ "trial_name": null,
357
+ "trial_params": null
358
+ }