emilys commited on
Commit
ee5640f
1 Parent(s): a34ae54
Files changed (1) hide show
  1. trainer_state.json +1204 -0
trainer_state.json ADDED
@@ -0,0 +1,1204 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6788413098236775,
3
+ "best_model_checkpoint": "trained_models/wnut_17/checkpoint-9000",
4
+ "epoch": 21.176470588235293,
5
+ "global_step": 9000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.24,
12
+ "eval_accuracy_score": 0.9205491641772071,
13
+ "eval_f1": 0.0,
14
+ "eval_loss": 0.5049907565116882,
15
+ "eval_precision": 0.0,
16
+ "eval_recall": 0.0,
17
+ "eval_runtime": 6.761,
18
+ "eval_samples_per_second": 149.239,
19
+ "eval_steps_per_second": 18.784,
20
+ "step": 100
21
+ },
22
+ {
23
+ "epoch": 0.47,
24
+ "eval_accuracy_score": 0.9205491641772071,
25
+ "eval_f1": 0.0,
26
+ "eval_loss": 0.48477333784103394,
27
+ "eval_precision": 0.0,
28
+ "eval_recall": 0.0,
29
+ "eval_runtime": 6.7798,
30
+ "eval_samples_per_second": 148.825,
31
+ "eval_steps_per_second": 18.732,
32
+ "step": 200
33
+ },
34
+ {
35
+ "epoch": 0.71,
36
+ "eval_accuracy_score": 0.9205491641772071,
37
+ "eval_f1": 0.0,
38
+ "eval_loss": 0.3834511935710907,
39
+ "eval_precision": 0.0,
40
+ "eval_recall": 0.0,
41
+ "eval_runtime": 6.7494,
42
+ "eval_samples_per_second": 149.494,
43
+ "eval_steps_per_second": 18.816,
44
+ "step": 300
45
+ },
46
+ {
47
+ "epoch": 0.94,
48
+ "eval_accuracy_score": 0.9218203775503718,
49
+ "eval_f1": 0.016241299303944315,
50
+ "eval_loss": 0.3428375720977783,
51
+ "eval_precision": 0.2692307692307692,
52
+ "eval_recall": 0.008373205741626795,
53
+ "eval_runtime": 6.7397,
54
+ "eval_samples_per_second": 149.711,
55
+ "eval_steps_per_second": 18.844,
56
+ "step": 400
57
+ },
58
+ {
59
+ "epoch": 1.18,
60
+ "learning_rate": 4.803921568627452e-06,
61
+ "loss": 0.4388,
62
+ "step": 500
63
+ },
64
+ {
65
+ "epoch": 1.18,
66
+ "eval_accuracy_score": 0.9387275154134621,
67
+ "eval_f1": 0.3550877192982456,
68
+ "eval_loss": 0.31836700439453125,
69
+ "eval_precision": 0.4295415959252971,
70
+ "eval_recall": 0.3026315789473684,
71
+ "eval_runtime": 6.749,
72
+ "eval_samples_per_second": 149.504,
73
+ "eval_steps_per_second": 18.818,
74
+ "step": 500
75
+ },
76
+ {
77
+ "epoch": 1.41,
78
+ "eval_accuracy_score": 0.9413335028284497,
79
+ "eval_f1": 0.40223463687150846,
80
+ "eval_loss": 0.3077085316181183,
81
+ "eval_precision": 0.48322147651006714,
82
+ "eval_recall": 0.3444976076555024,
83
+ "eval_runtime": 7.2416,
84
+ "eval_samples_per_second": 139.333,
85
+ "eval_steps_per_second": 17.537,
86
+ "step": 600
87
+ },
88
+ {
89
+ "epoch": 1.65,
90
+ "eval_accuracy_score": 0.9449564609419691,
91
+ "eval_f1": 0.46935933147632314,
92
+ "eval_loss": 0.29629671573638916,
93
+ "eval_precision": 0.5616666666666666,
94
+ "eval_recall": 0.40311004784689,
95
+ "eval_runtime": 6.7314,
96
+ "eval_samples_per_second": 149.895,
97
+ "eval_steps_per_second": 18.867,
98
+ "step": 700
99
+ },
100
+ {
101
+ "epoch": 1.88,
102
+ "eval_accuracy_score": 0.9471810843450074,
103
+ "eval_f1": 0.5139186295503213,
104
+ "eval_loss": 0.2875679135322571,
105
+ "eval_precision": 0.6371681415929203,
106
+ "eval_recall": 0.430622009569378,
107
+ "eval_runtime": 6.7537,
108
+ "eval_samples_per_second": 149.401,
109
+ "eval_steps_per_second": 18.805,
110
+ "step": 800
111
+ },
112
+ {
113
+ "epoch": 2.12,
114
+ "eval_accuracy_score": 0.949278586410729,
115
+ "eval_f1": 0.5329849771391247,
116
+ "eval_loss": 0.27145618200302124,
117
+ "eval_precision": 0.5870503597122302,
118
+ "eval_recall": 0.4880382775119617,
119
+ "eval_runtime": 6.7363,
120
+ "eval_samples_per_second": 149.786,
121
+ "eval_steps_per_second": 18.853,
122
+ "step": 900
123
+ },
124
+ {
125
+ "epoch": 2.35,
126
+ "learning_rate": 4.607843137254902e-06,
127
+ "loss": 0.1974,
128
+ "step": 1000
129
+ },
130
+ {
131
+ "epoch": 2.35,
132
+ "eval_accuracy_score": 0.9492150257420708,
133
+ "eval_f1": 0.5209424083769634,
134
+ "eval_loss": 0.2602783441543579,
135
+ "eval_precision": 0.5751445086705202,
136
+ "eval_recall": 0.47607655502392343,
137
+ "eval_runtime": 6.736,
138
+ "eval_samples_per_second": 149.792,
139
+ "eval_steps_per_second": 18.854,
140
+ "step": 1000
141
+ },
142
+ {
143
+ "epoch": 2.59,
144
+ "eval_accuracy_score": 0.9504862391152354,
145
+ "eval_f1": 0.5558510638297872,
146
+ "eval_loss": 0.24513839185237885,
147
+ "eval_precision": 0.625748502994012,
148
+ "eval_recall": 0.5,
149
+ "eval_runtime": 7.332,
150
+ "eval_samples_per_second": 137.616,
151
+ "eval_steps_per_second": 17.321,
152
+ "step": 1100
153
+ },
154
+ {
155
+ "epoch": 2.82,
156
+ "eval_accuracy_score": 0.9520116951630331,
157
+ "eval_f1": 0.5626566416040101,
158
+ "eval_loss": 0.23386946320533752,
159
+ "eval_precision": 0.5907894736842105,
160
+ "eval_recall": 0.5370813397129187,
161
+ "eval_runtime": 6.737,
162
+ "eval_samples_per_second": 149.771,
163
+ "eval_steps_per_second": 18.851,
164
+ "step": 1200
165
+ },
166
+ {
167
+ "epoch": 3.06,
168
+ "eval_accuracy_score": 0.9529651051929067,
169
+ "eval_f1": 0.5635782747603835,
170
+ "eval_loss": 0.23247480392456055,
171
+ "eval_precision": 0.6049382716049383,
172
+ "eval_recall": 0.527511961722488,
173
+ "eval_runtime": 6.7326,
174
+ "eval_samples_per_second": 149.868,
175
+ "eval_steps_per_second": 18.864,
176
+ "step": 1300
177
+ },
178
+ {
179
+ "epoch": 3.29,
180
+ "eval_accuracy_score": 0.9557617746138689,
181
+ "eval_f1": 0.5892526250772081,
182
+ "eval_loss": 0.2185893952846527,
183
+ "eval_precision": 0.6091954022988506,
184
+ "eval_recall": 0.5705741626794258,
185
+ "eval_runtime": 6.7427,
186
+ "eval_samples_per_second": 149.643,
187
+ "eval_steps_per_second": 18.835,
188
+ "step": 1400
189
+ },
190
+ {
191
+ "epoch": 3.53,
192
+ "learning_rate": 4.411764705882353e-06,
193
+ "loss": 0.1395,
194
+ "step": 1500
195
+ },
196
+ {
197
+ "epoch": 3.53,
198
+ "eval_accuracy_score": 0.9572236699930083,
199
+ "eval_f1": 0.6000000000000001,
200
+ "eval_loss": 0.2226746827363968,
201
+ "eval_precision": 0.6464088397790055,
202
+ "eval_recall": 0.5598086124401914,
203
+ "eval_runtime": 6.7181,
204
+ "eval_samples_per_second": 150.191,
205
+ "eval_steps_per_second": 18.904,
206
+ "step": 1500
207
+ },
208
+ {
209
+ "epoch": 3.76,
210
+ "eval_accuracy_score": 0.9590033687154389,
211
+ "eval_f1": 0.598405885959534,
212
+ "eval_loss": 0.21203213930130005,
213
+ "eval_precision": 0.6138364779874214,
214
+ "eval_recall": 0.583732057416268,
215
+ "eval_runtime": 6.7595,
216
+ "eval_samples_per_second": 149.27,
217
+ "eval_steps_per_second": 18.788,
218
+ "step": 1600
219
+ },
220
+ {
221
+ "epoch": 4.0,
222
+ "eval_accuracy_score": 0.9586855653721477,
223
+ "eval_f1": 0.618421052631579,
224
+ "eval_loss": 0.22468985617160797,
225
+ "eval_precision": 0.6871345029239766,
226
+ "eval_recall": 0.562200956937799,
227
+ "eval_runtime": 6.7215,
228
+ "eval_samples_per_second": 150.115,
229
+ "eval_steps_per_second": 18.895,
230
+ "step": 1700
231
+ },
232
+ {
233
+ "epoch": 4.24,
234
+ "eval_accuracy_score": 0.9599567787453124,
235
+ "eval_f1": 0.6330749354005167,
236
+ "eval_loss": 0.2214759886264801,
237
+ "eval_precision": 0.6882022471910112,
238
+ "eval_recall": 0.5861244019138756,
239
+ "eval_runtime": 6.7281,
240
+ "eval_samples_per_second": 149.969,
241
+ "eval_steps_per_second": 18.876,
242
+ "step": 1800
243
+ },
244
+ {
245
+ "epoch": 4.47,
246
+ "eval_accuracy_score": 0.96148223479311,
247
+ "eval_f1": 0.6507115135834413,
248
+ "eval_loss": 0.21751971542835236,
249
+ "eval_precision": 0.7084507042253522,
250
+ "eval_recall": 0.6016746411483254,
251
+ "eval_runtime": 6.7505,
252
+ "eval_samples_per_second": 149.471,
253
+ "eval_steps_per_second": 18.813,
254
+ "step": 1900
255
+ },
256
+ {
257
+ "epoch": 4.71,
258
+ "learning_rate": 4.215686274509805e-06,
259
+ "loss": 0.1057,
260
+ "step": 2000
261
+ },
262
+ {
263
+ "epoch": 4.71,
264
+ "eval_accuracy_score": 0.9605923854318947,
265
+ "eval_f1": 0.6401062416998671,
266
+ "eval_loss": 0.22828274965286255,
267
+ "eval_precision": 0.7194029850746269,
268
+ "eval_recall": 0.5765550239234449,
269
+ "eval_runtime": 6.7552,
270
+ "eval_samples_per_second": 149.366,
271
+ "eval_steps_per_second": 18.8,
272
+ "step": 2000
273
+ },
274
+ {
275
+ "epoch": 4.94,
276
+ "eval_accuracy_score": 0.9621814021483506,
277
+ "eval_f1": 0.6484424666242848,
278
+ "eval_loss": 0.2120533287525177,
279
+ "eval_precision": 0.6919945725915875,
280
+ "eval_recall": 0.6100478468899522,
281
+ "eval_runtime": 6.7607,
282
+ "eval_samples_per_second": 149.245,
283
+ "eval_steps_per_second": 18.785,
284
+ "step": 2100
285
+ },
286
+ {
287
+ "epoch": 5.18,
288
+ "eval_accuracy_score": 0.9617364774677429,
289
+ "eval_f1": 0.6427196921103272,
290
+ "eval_loss": 0.21272720396518707,
291
+ "eval_precision": 0.6929460580912863,
292
+ "eval_recall": 0.5992822966507177,
293
+ "eval_runtime": 6.823,
294
+ "eval_samples_per_second": 147.882,
295
+ "eval_steps_per_second": 18.613,
296
+ "step": 2200
297
+ },
298
+ {
299
+ "epoch": 5.41,
300
+ "eval_accuracy_score": 0.9614186741244518,
301
+ "eval_f1": 0.6440460947503202,
302
+ "eval_loss": 0.21506403386592865,
303
+ "eval_precision": 0.6928374655647382,
304
+ "eval_recall": 0.6016746411483254,
305
+ "eval_runtime": 6.7509,
306
+ "eval_samples_per_second": 149.462,
307
+ "eval_steps_per_second": 18.812,
308
+ "step": 2300
309
+ },
310
+ {
311
+ "epoch": 5.65,
312
+ "eval_accuracy_score": 0.9625627661603,
313
+ "eval_f1": 0.6498740554156172,
314
+ "eval_loss": 0.20958468317985535,
315
+ "eval_precision": 0.6861702127659575,
316
+ "eval_recall": 0.6172248803827751,
317
+ "eval_runtime": 6.7421,
318
+ "eval_samples_per_second": 149.656,
319
+ "eval_steps_per_second": 18.837,
320
+ "step": 2400
321
+ },
322
+ {
323
+ "epoch": 5.88,
324
+ "learning_rate": 4.019607843137255e-06,
325
+ "loss": 0.0811,
326
+ "step": 2500
327
+ },
328
+ {
329
+ "epoch": 5.88,
330
+ "eval_accuracy_score": 0.9624356448229835,
331
+ "eval_f1": 0.6591337099811676,
332
+ "eval_loss": 0.2008962631225586,
333
+ "eval_precision": 0.6935270805812418,
334
+ "eval_recall": 0.6279904306220095,
335
+ "eval_runtime": 6.7089,
336
+ "eval_samples_per_second": 150.397,
337
+ "eval_steps_per_second": 18.93,
338
+ "step": 2500
339
+ },
340
+ {
341
+ "epoch": 6.12,
342
+ "eval_accuracy_score": 0.9621178414796924,
343
+ "eval_f1": 0.6448717948717948,
344
+ "eval_loss": 0.21091684699058533,
345
+ "eval_precision": 0.6947513812154696,
346
+ "eval_recall": 0.6016746411483254,
347
+ "eval_runtime": 6.7609,
348
+ "eval_samples_per_second": 149.241,
349
+ "eval_steps_per_second": 18.785,
350
+ "step": 2600
351
+ },
352
+ {
353
+ "epoch": 6.35,
354
+ "eval_accuracy_score": 0.9621178414796924,
355
+ "eval_f1": 0.6499057196731616,
356
+ "eval_loss": 0.21349507570266724,
357
+ "eval_precision": 0.6847682119205298,
358
+ "eval_recall": 0.618421052631579,
359
+ "eval_runtime": 6.7376,
360
+ "eval_samples_per_second": 149.757,
361
+ "eval_steps_per_second": 18.849,
362
+ "step": 2700
363
+ },
364
+ {
365
+ "epoch": 6.59,
366
+ "eval_accuracy_score": 0.9620542808110342,
367
+ "eval_f1": 0.6467218332272437,
368
+ "eval_loss": 0.2102096974849701,
369
+ "eval_precision": 0.691156462585034,
370
+ "eval_recall": 0.6076555023923444,
371
+ "eval_runtime": 6.7394,
372
+ "eval_samples_per_second": 149.717,
373
+ "eval_steps_per_second": 18.844,
374
+ "step": 2800
375
+ },
376
+ {
377
+ "epoch": 6.82,
378
+ "eval_accuracy_score": 0.9630076908409076,
379
+ "eval_f1": 0.6658385093167701,
380
+ "eval_loss": 0.2015107125043869,
381
+ "eval_precision": 0.6925064599483204,
382
+ "eval_recall": 0.6411483253588517,
383
+ "eval_runtime": 6.7773,
384
+ "eval_samples_per_second": 148.879,
385
+ "eval_steps_per_second": 18.739,
386
+ "step": 2900
387
+ },
388
+ {
389
+ "epoch": 7.06,
390
+ "learning_rate": 3.8235294117647055e-06,
391
+ "loss": 0.0674,
392
+ "step": 3000
393
+ },
394
+ {
395
+ "epoch": 7.06,
396
+ "eval_accuracy_score": 0.96364329752749,
397
+ "eval_f1": 0.6624921728240452,
398
+ "eval_loss": 0.20179474353790283,
399
+ "eval_precision": 0.695137976346912,
400
+ "eval_recall": 0.6327751196172249,
401
+ "eval_runtime": 6.8013,
402
+ "eval_samples_per_second": 148.354,
403
+ "eval_steps_per_second": 18.673,
404
+ "step": 3000
405
+ },
406
+ {
407
+ "epoch": 7.29,
408
+ "eval_accuracy_score": 0.9629441301722494,
409
+ "eval_f1": 0.6454431960049937,
410
+ "eval_loss": 0.20865464210510254,
411
+ "eval_precision": 0.674934725848564,
412
+ "eval_recall": 0.618421052631579,
413
+ "eval_runtime": 6.7901,
414
+ "eval_samples_per_second": 148.599,
415
+ "eval_steps_per_second": 18.704,
416
+ "step": 3100
417
+ },
418
+ {
419
+ "epoch": 7.53,
420
+ "eval_accuracy_score": 0.9631983728468824,
421
+ "eval_f1": 0.6507237256135934,
422
+ "eval_loss": 0.2077988088130951,
423
+ "eval_precision": 0.6865869853917662,
424
+ "eval_recall": 0.618421052631579,
425
+ "eval_runtime": 6.7566,
426
+ "eval_samples_per_second": 149.335,
427
+ "eval_steps_per_second": 18.796,
428
+ "step": 3200
429
+ },
430
+ {
431
+ "epoch": 7.76,
432
+ "eval_accuracy_score": 0.9626898874976165,
433
+ "eval_f1": 0.6499678869621066,
434
+ "eval_loss": 0.2155311405658722,
435
+ "eval_precision": 0.7018030513176144,
436
+ "eval_recall": 0.6052631578947368,
437
+ "eval_runtime": 6.7572,
438
+ "eval_samples_per_second": 149.321,
439
+ "eval_steps_per_second": 18.795,
440
+ "step": 3300
441
+ },
442
+ {
443
+ "epoch": 8.0,
444
+ "eval_accuracy_score": 0.9624356448229835,
445
+ "eval_f1": 0.6524271844660194,
446
+ "eval_loss": 0.21822784841060638,
447
+ "eval_precision": 0.7108603667136812,
448
+ "eval_recall": 0.6028708133971292,
449
+ "eval_runtime": 6.7807,
450
+ "eval_samples_per_second": 148.804,
451
+ "eval_steps_per_second": 18.73,
452
+ "step": 3400
453
+ },
454
+ {
455
+ "epoch": 8.24,
456
+ "learning_rate": 3.6274509803921573e-06,
457
+ "loss": 0.0559,
458
+ "step": 3500
459
+ },
460
+ {
461
+ "epoch": 8.24,
462
+ "eval_accuracy_score": 0.9631348121782242,
463
+ "eval_f1": 0.659062103929024,
464
+ "eval_loss": 0.21273677051067352,
465
+ "eval_precision": 0.7008086253369272,
466
+ "eval_recall": 0.6220095693779905,
467
+ "eval_runtime": 6.7444,
468
+ "eval_samples_per_second": 149.606,
469
+ "eval_steps_per_second": 18.831,
470
+ "step": 3500
471
+ },
472
+ {
473
+ "epoch": 8.47,
474
+ "eval_accuracy_score": 0.9634526155215153,
475
+ "eval_f1": 0.6608478802992518,
476
+ "eval_loss": 0.20193149149417877,
477
+ "eval_precision": 0.6901041666666666,
478
+ "eval_recall": 0.6339712918660287,
479
+ "eval_runtime": 8.1777,
480
+ "eval_samples_per_second": 123.384,
481
+ "eval_steps_per_second": 15.53,
482
+ "step": 3600
483
+ },
484
+ {
485
+ "epoch": 8.71,
486
+ "eval_accuracy_score": 0.9634526155215153,
487
+ "eval_f1": 0.6587942821628341,
488
+ "eval_loss": 0.20160645246505737,
489
+ "eval_precision": 0.685640362225097,
490
+ "eval_recall": 0.6339712918660287,
491
+ "eval_runtime": 6.7668,
492
+ "eval_samples_per_second": 149.11,
493
+ "eval_steps_per_second": 18.768,
494
+ "step": 3700
495
+ },
496
+ {
497
+ "epoch": 8.94,
498
+ "eval_accuracy_score": 0.9634526155215153,
499
+ "eval_f1": 0.662553979025293,
500
+ "eval_loss": 0.20665320754051208,
501
+ "eval_precision": 0.6840764331210191,
502
+ "eval_recall": 0.6423444976076556,
503
+ "eval_runtime": 6.7328,
504
+ "eval_samples_per_second": 149.863,
505
+ "eval_steps_per_second": 18.863,
506
+ "step": 3800
507
+ },
508
+ {
509
+ "epoch": 9.18,
510
+ "eval_accuracy_score": 0.9629441301722494,
511
+ "eval_f1": 0.6566775244299674,
512
+ "eval_loss": 0.22287048399448395,
513
+ "eval_precision": 0.721030042918455,
514
+ "eval_recall": 0.6028708133971292,
515
+ "eval_runtime": 6.7851,
516
+ "eval_samples_per_second": 148.709,
517
+ "eval_steps_per_second": 18.718,
518
+ "step": 3900
519
+ },
520
+ {
521
+ "epoch": 9.41,
522
+ "learning_rate": 3.431372549019608e-06,
523
+ "loss": 0.0459,
524
+ "step": 4000
525
+ },
526
+ {
527
+ "epoch": 9.41,
528
+ "eval_accuracy_score": 0.9635797368588318,
529
+ "eval_f1": 0.6574827802128992,
530
+ "eval_loss": 0.20987075567245483,
531
+ "eval_precision": 0.6898817345597897,
532
+ "eval_recall": 0.6279904306220095,
533
+ "eval_runtime": 6.7469,
534
+ "eval_samples_per_second": 149.549,
535
+ "eval_steps_per_second": 18.823,
536
+ "step": 4000
537
+ },
538
+ {
539
+ "epoch": 9.65,
540
+ "eval_accuracy_score": 0.9628805695035911,
541
+ "eval_f1": 0.6568690095846645,
542
+ "eval_loss": 0.22190582752227783,
543
+ "eval_precision": 0.7050754458161865,
544
+ "eval_recall": 0.6148325358851675,
545
+ "eval_runtime": 6.7586,
546
+ "eval_samples_per_second": 149.292,
547
+ "eval_steps_per_second": 18.791,
548
+ "step": 4100
549
+ },
550
+ {
551
+ "epoch": 9.88,
552
+ "eval_accuracy_score": 0.9632619335155406,
553
+ "eval_f1": 0.6525265127885215,
554
+ "eval_loss": 0.21028834581375122,
555
+ "eval_precision": 0.681877444589309,
556
+ "eval_recall": 0.6255980861244019,
557
+ "eval_runtime": 6.7583,
558
+ "eval_samples_per_second": 149.298,
559
+ "eval_steps_per_second": 18.792,
560
+ "step": 4200
561
+ },
562
+ {
563
+ "epoch": 10.12,
564
+ "eval_accuracy_score": 0.9640882222080976,
565
+ "eval_f1": 0.6692111959287532,
566
+ "eval_loss": 0.2170829027891159,
567
+ "eval_precision": 0.7146739130434783,
568
+ "eval_recall": 0.6291866028708134,
569
+ "eval_runtime": 6.7544,
570
+ "eval_samples_per_second": 149.383,
571
+ "eval_steps_per_second": 18.802,
572
+ "step": 4300
573
+ },
574
+ {
575
+ "epoch": 10.35,
576
+ "eval_accuracy_score": 0.9630712515095659,
577
+ "eval_f1": 0.6495619524405507,
578
+ "eval_loss": 0.2193835973739624,
579
+ "eval_precision": 0.6811023622047244,
580
+ "eval_recall": 0.6208133971291866,
581
+ "eval_runtime": 6.7362,
582
+ "eval_samples_per_second": 149.787,
583
+ "eval_steps_per_second": 18.853,
584
+ "step": 4400
585
+ },
586
+ {
587
+ "epoch": 10.59,
588
+ "learning_rate": 3.2352941176470594e-06,
589
+ "loss": 0.0399,
590
+ "step": 4500
591
+ },
592
+ {
593
+ "epoch": 10.59,
594
+ "eval_accuracy_score": 0.9634526155215153,
595
+ "eval_f1": 0.6556543837357053,
596
+ "eval_loss": 0.21863853931427002,
597
+ "eval_precision": 0.6991869918699187,
598
+ "eval_recall": 0.6172248803827751,
599
+ "eval_runtime": 6.769,
600
+ "eval_samples_per_second": 149.062,
601
+ "eval_steps_per_second": 18.762,
602
+ "step": 4500
603
+ },
604
+ {
605
+ "epoch": 10.82,
606
+ "eval_accuracy_score": 0.9637068581961482,
607
+ "eval_f1": 0.6624685138539044,
608
+ "eval_loss": 0.21445031464099884,
609
+ "eval_precision": 0.699468085106383,
610
+ "eval_recall": 0.6291866028708134,
611
+ "eval_runtime": 6.7298,
612
+ "eval_samples_per_second": 149.93,
613
+ "eval_steps_per_second": 18.871,
614
+ "step": 4600
615
+ },
616
+ {
617
+ "epoch": 11.06,
618
+ "eval_accuracy_score": 0.9644695862200471,
619
+ "eval_f1": 0.6759142496847416,
620
+ "eval_loss": 0.20991285145282745,
621
+ "eval_precision": 0.7146666666666667,
622
+ "eval_recall": 0.6411483253588517,
623
+ "eval_runtime": 6.734,
624
+ "eval_samples_per_second": 149.837,
625
+ "eval_steps_per_second": 18.86,
626
+ "step": 4700
627
+ },
628
+ {
629
+ "epoch": 11.29,
630
+ "eval_accuracy_score": 0.96364329752749,
631
+ "eval_f1": 0.6649778340721977,
632
+ "eval_loss": 0.22495105862617493,
633
+ "eval_precision": 0.7065948855989233,
634
+ "eval_recall": 0.6279904306220095,
635
+ "eval_runtime": 6.7408,
636
+ "eval_samples_per_second": 149.687,
637
+ "eval_steps_per_second": 18.841,
638
+ "step": 4800
639
+ },
640
+ {
641
+ "epoch": 11.53,
642
+ "eval_accuracy_score": 0.9639611008707811,
643
+ "eval_f1": 0.6670894102726697,
644
+ "eval_loss": 0.22267203032970428,
645
+ "eval_precision": 0.7098515519568152,
646
+ "eval_recall": 0.6291866028708134,
647
+ "eval_runtime": 6.7689,
648
+ "eval_samples_per_second": 149.064,
649
+ "eval_steps_per_second": 18.762,
650
+ "step": 4900
651
+ },
652
+ {
653
+ "epoch": 11.76,
654
+ "learning_rate": 3.03921568627451e-06,
655
+ "loss": 0.0368,
656
+ "step": 5000
657
+ },
658
+ {
659
+ "epoch": 11.76,
660
+ "eval_accuracy_score": 0.9637068581961482,
661
+ "eval_f1": 0.6588235294117647,
662
+ "eval_loss": 0.21161217987537384,
663
+ "eval_precision": 0.6829268292682927,
664
+ "eval_recall": 0.6363636363636364,
665
+ "eval_runtime": 6.7536,
666
+ "eval_samples_per_second": 149.402,
667
+ "eval_steps_per_second": 18.805,
668
+ "step": 5000
669
+ },
670
+ {
671
+ "epoch": 12.0,
672
+ "eval_accuracy_score": 0.9625627661603,
673
+ "eval_f1": 0.6593830334190232,
674
+ "eval_loss": 0.227540984749794,
675
+ "eval_precision": 0.7125,
676
+ "eval_recall": 0.6136363636363636,
677
+ "eval_runtime": 6.7584,
678
+ "eval_samples_per_second": 149.295,
679
+ "eval_steps_per_second": 18.791,
680
+ "step": 5100
681
+ },
682
+ {
683
+ "epoch": 12.24,
684
+ "eval_accuracy_score": 0.9628805695035911,
685
+ "eval_f1": 0.6640471512770137,
686
+ "eval_loss": 0.23519518971443176,
687
+ "eval_precision": 0.7337192474674384,
688
+ "eval_recall": 0.6064593301435407,
689
+ "eval_runtime": 6.7729,
690
+ "eval_samples_per_second": 148.976,
691
+ "eval_steps_per_second": 18.751,
692
+ "step": 5200
693
+ },
694
+ {
695
+ "epoch": 12.47,
696
+ "eval_accuracy_score": 0.9628170088349329,
697
+ "eval_f1": 0.6679462571976967,
698
+ "eval_loss": 0.2271033227443695,
699
+ "eval_precision": 0.7180192572214581,
700
+ "eval_recall": 0.6244019138755981,
701
+ "eval_runtime": 6.7726,
702
+ "eval_samples_per_second": 148.983,
703
+ "eval_steps_per_second": 18.752,
704
+ "step": 5300
705
+ },
706
+ {
707
+ "epoch": 12.71,
708
+ "eval_accuracy_score": 0.9630076908409076,
709
+ "eval_f1": 0.6628643852978454,
710
+ "eval_loss": 0.22853700816631317,
711
+ "eval_precision": 0.7048517520215634,
712
+ "eval_recall": 0.6255980861244019,
713
+ "eval_runtime": 6.7535,
714
+ "eval_samples_per_second": 149.405,
715
+ "eval_steps_per_second": 18.805,
716
+ "step": 5400
717
+ },
718
+ {
719
+ "epoch": 12.94,
720
+ "learning_rate": 2.843137254901961e-06,
721
+ "loss": 0.0321,
722
+ "step": 5500
723
+ },
724
+ {
725
+ "epoch": 12.94,
726
+ "eval_accuracy_score": 0.9631348121782242,
727
+ "eval_f1": 0.664548919949174,
728
+ "eval_loss": 0.22333784401416779,
729
+ "eval_precision": 0.7086720867208672,
730
+ "eval_recall": 0.6255980861244019,
731
+ "eval_runtime": 6.7401,
732
+ "eval_samples_per_second": 149.702,
733
+ "eval_steps_per_second": 18.843,
734
+ "step": 5500
735
+ },
736
+ {
737
+ "epoch": 13.18,
738
+ "eval_accuracy_score": 0.9637068581961482,
739
+ "eval_f1": 0.6705054382597568,
740
+ "eval_loss": 0.222885861992836,
741
+ "eval_precision": 0.7207702888583218,
742
+ "eval_recall": 0.6267942583732058,
743
+ "eval_runtime": 6.7486,
744
+ "eval_samples_per_second": 149.513,
745
+ "eval_steps_per_second": 18.819,
746
+ "step": 5600
747
+ },
748
+ {
749
+ "epoch": 13.41,
750
+ "eval_accuracy_score": 0.9633254941841988,
751
+ "eval_f1": 0.6619718309859155,
752
+ "eval_loss": 0.2285127490758896,
753
+ "eval_precision": 0.7121212121212122,
754
+ "eval_recall": 0.618421052631579,
755
+ "eval_runtime": 6.8129,
756
+ "eval_samples_per_second": 148.102,
757
+ "eval_steps_per_second": 18.641,
758
+ "step": 5700
759
+ },
760
+ {
761
+ "epoch": 13.65,
762
+ "eval_accuracy_score": 0.9644695862200471,
763
+ "eval_f1": 0.6734177215189874,
764
+ "eval_loss": 0.22083869576454163,
765
+ "eval_precision": 0.7150537634408602,
766
+ "eval_recall": 0.6363636363636364,
767
+ "eval_runtime": 6.7283,
768
+ "eval_samples_per_second": 149.964,
769
+ "eval_steps_per_second": 18.876,
770
+ "step": 5800
771
+ },
772
+ {
773
+ "epoch": 13.88,
774
+ "eval_accuracy_score": 0.9635161761901735,
775
+ "eval_f1": 0.6684005201560469,
776
+ "eval_loss": 0.2334665209054947,
777
+ "eval_precision": 0.7321937321937322,
778
+ "eval_recall": 0.6148325358851675,
779
+ "eval_runtime": 6.7215,
780
+ "eval_samples_per_second": 150.116,
781
+ "eval_steps_per_second": 18.895,
782
+ "step": 5900
783
+ },
784
+ {
785
+ "epoch": 14.12,
786
+ "learning_rate": 2.647058823529412e-06,
787
+ "loss": 0.0291,
788
+ "step": 6000
789
+ },
790
+ {
791
+ "epoch": 14.12,
792
+ "eval_accuracy_score": 0.9641517828767558,
793
+ "eval_f1": 0.663265306122449,
794
+ "eval_loss": 0.2206926792860031,
795
+ "eval_precision": 0.7103825136612022,
796
+ "eval_recall": 0.6220095693779905,
797
+ "eval_runtime": 6.7314,
798
+ "eval_samples_per_second": 149.894,
799
+ "eval_steps_per_second": 18.867,
800
+ "step": 6000
801
+ },
802
+ {
803
+ "epoch": 14.35,
804
+ "eval_accuracy_score": 0.9650416322379711,
805
+ "eval_f1": 0.6737633061991233,
806
+ "eval_loss": 0.21664156019687653,
807
+ "eval_precision": 0.7069645203679369,
808
+ "eval_recall": 0.6435406698564593,
809
+ "eval_runtime": 6.8128,
810
+ "eval_samples_per_second": 148.104,
811
+ "eval_steps_per_second": 18.641,
812
+ "step": 6100
813
+ },
814
+ {
815
+ "epoch": 14.59,
816
+ "eval_accuracy_score": 0.9646602682260217,
817
+ "eval_f1": 0.6654275092936804,
818
+ "eval_loss": 0.20899562537670135,
819
+ "eval_precision": 0.6902313624678663,
820
+ "eval_recall": 0.6423444976076556,
821
+ "eval_runtime": 6.7812,
822
+ "eval_samples_per_second": 148.794,
823
+ "eval_steps_per_second": 18.728,
824
+ "step": 6200
825
+ },
826
+ {
827
+ "epoch": 14.82,
828
+ "eval_accuracy_score": 0.9647873895633382,
829
+ "eval_f1": 0.6766917293233083,
830
+ "eval_loss": 0.21245865523815155,
831
+ "eval_precision": 0.7105263157894737,
832
+ "eval_recall": 0.645933014354067,
833
+ "eval_runtime": 6.7236,
834
+ "eval_samples_per_second": 150.069,
835
+ "eval_steps_per_second": 18.889,
836
+ "step": 6300
837
+ },
838
+ {
839
+ "epoch": 15.06,
840
+ "eval_accuracy_score": 0.9637704188648064,
841
+ "eval_f1": 0.6666666666666667,
842
+ "eval_loss": 0.22959989309310913,
843
+ "eval_precision": 0.713506139154161,
844
+ "eval_recall": 0.6255980861244019,
845
+ "eval_runtime": 6.7691,
846
+ "eval_samples_per_second": 149.061,
847
+ "eval_steps_per_second": 18.762,
848
+ "step": 6400
849
+ },
850
+ {
851
+ "epoch": 15.29,
852
+ "learning_rate": 2.450980392156863e-06,
853
+ "loss": 0.0251,
854
+ "step": 6500
855
+ },
856
+ {
857
+ "epoch": 15.29,
858
+ "eval_accuracy_score": 0.9645331468887053,
859
+ "eval_f1": 0.6733542319749217,
860
+ "eval_loss": 0.22114437818527222,
861
+ "eval_precision": 0.7075098814229249,
862
+ "eval_recall": 0.6423444976076556,
863
+ "eval_runtime": 6.8495,
864
+ "eval_samples_per_second": 147.311,
865
+ "eval_steps_per_second": 18.542,
866
+ "step": 6500
867
+ },
868
+ {
869
+ "epoch": 15.53,
870
+ "eval_accuracy_score": 0.9640882222080976,
871
+ "eval_f1": 0.6713197969543148,
872
+ "eval_loss": 0.22384566068649292,
873
+ "eval_precision": 0.7148648648648649,
874
+ "eval_recall": 0.6327751196172249,
875
+ "eval_runtime": 6.7143,
876
+ "eval_samples_per_second": 150.277,
877
+ "eval_steps_per_second": 18.915,
878
+ "step": 6600
879
+ },
880
+ {
881
+ "epoch": 15.76,
882
+ "eval_accuracy_score": 0.9637068581961482,
883
+ "eval_f1": 0.6641172721478649,
884
+ "eval_loss": 0.23098178207874298,
885
+ "eval_precision": 0.7107776261937244,
886
+ "eval_recall": 0.6232057416267942,
887
+ "eval_runtime": 6.7457,
888
+ "eval_samples_per_second": 149.576,
889
+ "eval_steps_per_second": 18.827,
890
+ "step": 6700
891
+ },
892
+ {
893
+ "epoch": 16.0,
894
+ "eval_accuracy_score": 0.9650416322379711,
895
+ "eval_f1": 0.6762860727728984,
896
+ "eval_loss": 0.21991874277591705,
897
+ "eval_precision": 0.7110817941952506,
898
+ "eval_recall": 0.6447368421052632,
899
+ "eval_runtime": 6.7686,
900
+ "eval_samples_per_second": 149.07,
901
+ "eval_steps_per_second": 18.763,
902
+ "step": 6800
903
+ },
904
+ {
905
+ "epoch": 16.24,
906
+ "eval_accuracy_score": 0.96364329752749,
907
+ "eval_f1": 0.6636771300448431,
908
+ "eval_loss": 0.23369288444519043,
909
+ "eval_precision": 0.7144827586206897,
910
+ "eval_recall": 0.6196172248803827,
911
+ "eval_runtime": 6.8164,
912
+ "eval_samples_per_second": 148.025,
913
+ "eval_steps_per_second": 18.631,
914
+ "step": 6900
915
+ },
916
+ {
917
+ "epoch": 16.47,
918
+ "learning_rate": 2.254901960784314e-06,
919
+ "loss": 0.0227,
920
+ "step": 7000
921
+ },
922
+ {
923
+ "epoch": 16.47,
924
+ "eval_accuracy_score": 0.9646602682260217,
925
+ "eval_f1": 0.6712243629583592,
926
+ "eval_loss": 0.22149834036827087,
927
+ "eval_precision": 0.6985769728331177,
928
+ "eval_recall": 0.645933014354067,
929
+ "eval_runtime": 6.7644,
930
+ "eval_samples_per_second": 149.164,
931
+ "eval_steps_per_second": 18.775,
932
+ "step": 7000
933
+ },
934
+ {
935
+ "epoch": 16.71,
936
+ "eval_accuracy_score": 0.9645331468887053,
937
+ "eval_f1": 0.6687422166874221,
938
+ "eval_loss": 0.2212422788143158,
939
+ "eval_precision": 0.6974025974025974,
940
+ "eval_recall": 0.6423444976076556,
941
+ "eval_runtime": 6.7009,
942
+ "eval_samples_per_second": 150.576,
943
+ "eval_steps_per_second": 18.953,
944
+ "step": 7100
945
+ },
946
+ {
947
+ "epoch": 16.94,
948
+ "eval_accuracy_score": 0.9635161761901735,
949
+ "eval_f1": 0.6653821451509313,
950
+ "eval_loss": 0.24178394675254822,
951
+ "eval_precision": 0.7184466019417476,
952
+ "eval_recall": 0.6196172248803827,
953
+ "eval_runtime": 6.7952,
954
+ "eval_samples_per_second": 148.486,
955
+ "eval_steps_per_second": 18.69,
956
+ "step": 7200
957
+ },
958
+ {
959
+ "epoch": 17.18,
960
+ "eval_accuracy_score": 0.9633890548528571,
961
+ "eval_f1": 0.6675307841866494,
962
+ "eval_loss": 0.24713467061519623,
963
+ "eval_precision": 0.7284299858557284,
964
+ "eval_recall": 0.6160287081339713,
965
+ "eval_runtime": 6.807,
966
+ "eval_samples_per_second": 148.229,
967
+ "eval_steps_per_second": 18.657,
968
+ "step": 7300
969
+ },
970
+ {
971
+ "epoch": 17.41,
972
+ "eval_accuracy_score": 0.9643424648827306,
973
+ "eval_f1": 0.670031545741325,
974
+ "eval_loss": 0.22886203229427338,
975
+ "eval_precision": 0.7089452603471295,
976
+ "eval_recall": 0.6351674641148325,
977
+ "eval_runtime": 6.7775,
978
+ "eval_samples_per_second": 148.874,
979
+ "eval_steps_per_second": 18.738,
980
+ "step": 7400
981
+ },
982
+ {
983
+ "epoch": 17.65,
984
+ "learning_rate": 2.058823529411765e-06,
985
+ "loss": 0.0197,
986
+ "step": 7500
987
+ },
988
+ {
989
+ "epoch": 17.65,
990
+ "eval_accuracy_score": 0.9646602682260217,
991
+ "eval_f1": 0.6753731343283582,
992
+ "eval_loss": 0.22524945437908173,
993
+ "eval_precision": 0.7033678756476683,
994
+ "eval_recall": 0.6495215311004785,
995
+ "eval_runtime": 6.7338,
996
+ "eval_samples_per_second": 149.84,
997
+ "eval_steps_per_second": 18.86,
998
+ "step": 7500
999
+ },
1000
+ {
1001
+ "epoch": 17.88,
1002
+ "eval_accuracy_score": 0.9633890548528571,
1003
+ "eval_f1": 0.664528543938422,
1004
+ "eval_loss": 0.24701638519763947,
1005
+ "eval_precision": 0.7164591977869986,
1006
+ "eval_recall": 0.6196172248803827,
1007
+ "eval_runtime": 6.7838,
1008
+ "eval_samples_per_second": 148.736,
1009
+ "eval_steps_per_second": 18.721,
1010
+ "step": 7600
1011
+ },
1012
+ {
1013
+ "epoch": 18.12,
1014
+ "eval_accuracy_score": 0.9654865569185788,
1015
+ "eval_f1": 0.6825,
1016
+ "eval_loss": 0.22353364527225494,
1017
+ "eval_precision": 0.7146596858638743,
1018
+ "eval_recall": 0.65311004784689,
1019
+ "eval_runtime": 6.7595,
1020
+ "eval_samples_per_second": 149.272,
1021
+ "eval_steps_per_second": 18.788,
1022
+ "step": 7700
1023
+ },
1024
+ {
1025
+ "epoch": 18.35,
1026
+ "eval_accuracy_score": 0.9638975402021229,
1027
+ "eval_f1": 0.6590765338393421,
1028
+ "eval_loss": 0.23786970973014832,
1029
+ "eval_precision": 0.6993288590604027,
1030
+ "eval_recall": 0.6232057416267942,
1031
+ "eval_runtime": 6.8141,
1032
+ "eval_samples_per_second": 148.075,
1033
+ "eval_steps_per_second": 18.638,
1034
+ "step": 7800
1035
+ },
1036
+ {
1037
+ "epoch": 18.59,
1038
+ "eval_accuracy_score": 0.9644060255513888,
1039
+ "eval_f1": 0.6666666666666667,
1040
+ "eval_loss": 0.23264023661613464,
1041
+ "eval_precision": 0.7014531043593131,
1042
+ "eval_recall": 0.6351674641148325,
1043
+ "eval_runtime": 6.8013,
1044
+ "eval_samples_per_second": 148.353,
1045
+ "eval_steps_per_second": 18.673,
1046
+ "step": 7900
1047
+ },
1048
+ {
1049
+ "epoch": 18.82,
1050
+ "learning_rate": 1.8627450980392158e-06,
1051
+ "loss": 0.0183,
1052
+ "step": 8000
1053
+ },
1054
+ {
1055
+ "epoch": 18.82,
1056
+ "eval_accuracy_score": 0.9649780715693129,
1057
+ "eval_f1": 0.6724244293645897,
1058
+ "eval_loss": 0.22304123640060425,
1059
+ "eval_precision": 0.6942675159235668,
1060
+ "eval_recall": 0.6519138755980861,
1061
+ "eval_runtime": 6.7426,
1062
+ "eval_samples_per_second": 149.646,
1063
+ "eval_steps_per_second": 18.835,
1064
+ "step": 8000
1065
+ },
1066
+ {
1067
+ "epoch": 19.06,
1068
+ "eval_accuracy_score": 0.9643424648827306,
1069
+ "eval_f1": 0.6683480453972257,
1070
+ "eval_loss": 0.2350449413061142,
1071
+ "eval_precision": 0.7066666666666667,
1072
+ "eval_recall": 0.6339712918660287,
1073
+ "eval_runtime": 6.8057,
1074
+ "eval_samples_per_second": 148.259,
1075
+ "eval_steps_per_second": 18.661,
1076
+ "step": 8100
1077
+ },
1078
+ {
1079
+ "epoch": 19.29,
1080
+ "eval_accuracy_score": 0.9652323142439458,
1081
+ "eval_f1": 0.6790582403965303,
1082
+ "eval_loss": 0.226871520280838,
1083
+ "eval_precision": 0.7043701799485861,
1084
+ "eval_recall": 0.6555023923444976,
1085
+ "eval_runtime": 6.7483,
1086
+ "eval_samples_per_second": 149.519,
1087
+ "eval_steps_per_second": 18.819,
1088
+ "step": 8200
1089
+ },
1090
+ {
1091
+ "epoch": 19.53,
1092
+ "eval_accuracy_score": 0.96472382889468,
1093
+ "eval_f1": 0.6683354192740926,
1094
+ "eval_loss": 0.23186342418193817,
1095
+ "eval_precision": 0.7007874015748031,
1096
+ "eval_recall": 0.638755980861244,
1097
+ "eval_runtime": 6.7632,
1098
+ "eval_samples_per_second": 149.19,
1099
+ "eval_steps_per_second": 18.778,
1100
+ "step": 8300
1101
+ },
1102
+ {
1103
+ "epoch": 19.76,
1104
+ "eval_accuracy_score": 0.9645967075573635,
1105
+ "eval_f1": 0.6712933753943218,
1106
+ "eval_loss": 0.2337142825126648,
1107
+ "eval_precision": 0.7102803738317757,
1108
+ "eval_recall": 0.6363636363636364,
1109
+ "eval_runtime": 6.7451,
1110
+ "eval_samples_per_second": 149.591,
1111
+ "eval_steps_per_second": 18.829,
1112
+ "step": 8400
1113
+ },
1114
+ {
1115
+ "epoch": 20.0,
1116
+ "learning_rate": 1.6666666666666667e-06,
1117
+ "loss": 0.0162,
1118
+ "step": 8500
1119
+ },
1120
+ {
1121
+ "epoch": 20.0,
1122
+ "eval_accuracy_score": 0.9647873895633382,
1123
+ "eval_f1": 0.6751430387794024,
1124
+ "eval_loss": 0.2389359027147293,
1125
+ "eval_precision": 0.7204884667571235,
1126
+ "eval_recall": 0.6351674641148325,
1127
+ "eval_runtime": 6.7503,
1128
+ "eval_samples_per_second": 149.474,
1129
+ "eval_steps_per_second": 18.814,
1130
+ "step": 8500
1131
+ },
1132
+ {
1133
+ "epoch": 20.24,
1134
+ "eval_accuracy_score": 0.9638975402021229,
1135
+ "eval_f1": 0.6709677419354839,
1136
+ "eval_loss": 0.25359493494033813,
1137
+ "eval_precision": 0.7282913165266106,
1138
+ "eval_recall": 0.6220095693779905,
1139
+ "eval_runtime": 6.7422,
1140
+ "eval_samples_per_second": 149.655,
1141
+ "eval_steps_per_second": 18.837,
1142
+ "step": 8600
1143
+ },
1144
+ {
1145
+ "epoch": 20.47,
1146
+ "eval_accuracy_score": 0.964215343545414,
1147
+ "eval_f1": 0.6709346991037132,
1148
+ "eval_loss": 0.24693100154399872,
1149
+ "eval_precision": 0.721763085399449,
1150
+ "eval_recall": 0.6267942583732058,
1151
+ "eval_runtime": 6.7296,
1152
+ "eval_samples_per_second": 149.934,
1153
+ "eval_steps_per_second": 18.872,
1154
+ "step": 8700
1155
+ },
1156
+ {
1157
+ "epoch": 20.71,
1158
+ "eval_accuracy_score": 0.9637704188648064,
1159
+ "eval_f1": 0.6687938736439056,
1160
+ "eval_loss": 0.2458009421825409,
1161
+ "eval_precision": 0.7168262653898769,
1162
+ "eval_recall": 0.6267942583732058,
1163
+ "eval_runtime": 6.7627,
1164
+ "eval_samples_per_second": 149.201,
1165
+ "eval_steps_per_second": 18.78,
1166
+ "step": 8800
1167
+ },
1168
+ {
1169
+ "epoch": 20.94,
1170
+ "eval_accuracy_score": 0.9648509502319964,
1171
+ "eval_f1": 0.676304211187932,
1172
+ "eval_loss": 0.23430827260017395,
1173
+ "eval_precision": 0.7125827814569536,
1174
+ "eval_recall": 0.6435406698564593,
1175
+ "eval_runtime": 6.7289,
1176
+ "eval_samples_per_second": 149.951,
1177
+ "eval_steps_per_second": 18.874,
1178
+ "step": 8900
1179
+ },
1180
+ {
1181
+ "epoch": 21.18,
1182
+ "learning_rate": 1.4705882352941177e-06,
1183
+ "loss": 0.0149,
1184
+ "step": 9000
1185
+ },
1186
+ {
1187
+ "epoch": 21.18,
1188
+ "eval_accuracy_score": 0.9651687535752876,
1189
+ "eval_f1": 0.6788413098236775,
1190
+ "eval_loss": 0.23078210651874542,
1191
+ "eval_precision": 0.7167553191489362,
1192
+ "eval_recall": 0.6447368421052632,
1193
+ "eval_runtime": 6.7181,
1194
+ "eval_samples_per_second": 150.19,
1195
+ "eval_steps_per_second": 18.904,
1196
+ "step": 9000
1197
+ }
1198
+ ],
1199
+ "max_steps": 12750,
1200
+ "num_train_epochs": 30,
1201
+ "total_flos": 4695577703714304.0,
1202
+ "trial_name": null,
1203
+ "trial_params": null
1204
+ }