anoaky commited on
Commit
2f909ad
·
verified ·
1 Parent(s): 2051ea1

Training in progress, epoch 1, checkpoint

Browse files
checkpoint-1108/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google-t5/t5-large",
3
+ "architectures": [
4
+ "T5ForSequenceClassification"
5
+ ],
6
+ "classifier_dropout": 0.0,
7
+ "d_ff": 4096,
8
+ "d_kv": 64,
9
+ "d_model": 1024,
10
+ "decoder_start_token_id": 0,
11
+ "dense_act_fn": "relu",
12
+ "dropout_rate": 0.1,
13
+ "eos_token_id": 1,
14
+ "feed_forward_proj": "relu",
15
+ "initializer_factor": 1.0,
16
+ "is_encoder_decoder": true,
17
+ "is_gated_act": false,
18
+ "layer_norm_epsilon": 1e-06,
19
+ "model_type": "t5",
20
+ "n_positions": 512,
21
+ "num_decoder_layers": 24,
22
+ "num_heads": 16,
23
+ "num_layers": 24,
24
+ "output_past": true,
25
+ "pad_token_id": 0,
26
+ "relative_attention_max_distance": 128,
27
+ "relative_attention_num_buckets": 32,
28
+ "task_specific_params": {
29
+ "summarization": {
30
+ "early_stopping": true,
31
+ "length_penalty": 2.0,
32
+ "max_length": 200,
33
+ "min_length": 30,
34
+ "no_repeat_ngram_size": 3,
35
+ "num_beams": 4,
36
+ "prefix": "summarize: "
37
+ },
38
+ "translation_en_to_de": {
39
+ "early_stopping": true,
40
+ "max_length": 300,
41
+ "num_beams": 4,
42
+ "prefix": "translate English to German: "
43
+ },
44
+ "translation_en_to_fr": {
45
+ "early_stopping": true,
46
+ "max_length": 300,
47
+ "num_beams": 4,
48
+ "prefix": "translate English to French: "
49
+ },
50
+ "translation_en_to_ro": {
51
+ "early_stopping": true,
52
+ "max_length": 300,
53
+ "num_beams": 4,
54
+ "prefix": "translate English to Romanian: "
55
+ }
56
+ },
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.47.0",
59
+ "use_cache": true,
60
+ "vocab_size": 32128
61
+ }
checkpoint-1108/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:912b1f15c7ece3971722eaa44fab30a2bb743c9d1f76b43fed86e58ef8980ca1
3
+ size 2954947648
checkpoint-1108/trainer_state.json ADDED
@@ -0,0 +1,825 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.996845425867508,
5
+ "eval_steps": 500,
6
+ "global_step": 1108,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0,
13
+ "eval_f1": 0.0,
14
+ "eval_loss": 0.5020396709442139,
15
+ "eval_precision": 0.0,
16
+ "eval_recall": 0.0,
17
+ "eval_runtime": 132.3528,
18
+ "eval_samples_per_second": 35.307,
19
+ "eval_steps_per_second": 4.42,
20
+ "step": 0
21
+ },
22
+ {
23
+ "epoch": 0.018026137899954935,
24
+ "grad_norm": 11.620719909667969,
25
+ "learning_rate": 4.9548736462093865e-05,
26
+ "loss": 0.2661,
27
+ "step": 10
28
+ },
29
+ {
30
+ "epoch": 0.03605227579990987,
31
+ "grad_norm": 1.2112946510314941,
32
+ "learning_rate": 4.909747292418773e-05,
33
+ "loss": 0.0513,
34
+ "step": 20
35
+ },
36
+ {
37
+ "epoch": 0.054078413699864804,
38
+ "grad_norm": 0.3561903238296509,
39
+ "learning_rate": 4.864620938628159e-05,
40
+ "loss": 0.0542,
41
+ "step": 30
42
+ },
43
+ {
44
+ "epoch": 0.07210455159981974,
45
+ "grad_norm": 15.130791664123535,
46
+ "learning_rate": 4.819494584837546e-05,
47
+ "loss": 0.1113,
48
+ "step": 40
49
+ },
50
+ {
51
+ "epoch": 0.09013068949977468,
52
+ "grad_norm": 10.50532054901123,
53
+ "learning_rate": 4.7743682310469314e-05,
54
+ "loss": 0.0417,
55
+ "step": 50
56
+ },
57
+ {
58
+ "epoch": 0.10815682739972961,
59
+ "grad_norm": 7.872809410095215,
60
+ "learning_rate": 4.7292418772563177e-05,
61
+ "loss": 0.0652,
62
+ "step": 60
63
+ },
64
+ {
65
+ "epoch": 0.12618296529968454,
66
+ "grad_norm": 0.3549640476703644,
67
+ "learning_rate": 4.684115523465704e-05,
68
+ "loss": 0.0476,
69
+ "step": 70
70
+ },
71
+ {
72
+ "epoch": 0.14420910319963948,
73
+ "grad_norm": 3.7721316814422607,
74
+ "learning_rate": 4.63898916967509e-05,
75
+ "loss": 0.0489,
76
+ "step": 80
77
+ },
78
+ {
79
+ "epoch": 0.16223524109959442,
80
+ "grad_norm": 1.7026127576828003,
81
+ "learning_rate": 4.5938628158844764e-05,
82
+ "loss": 0.0795,
83
+ "step": 90
84
+ },
85
+ {
86
+ "epoch": 0.18026137899954936,
87
+ "grad_norm": 0.6303790807723999,
88
+ "learning_rate": 4.548736462093863e-05,
89
+ "loss": 0.0187,
90
+ "step": 100
91
+ },
92
+ {
93
+ "epoch": 0.19828751689950427,
94
+ "grad_norm": 0.14402388036251068,
95
+ "learning_rate": 4.5036101083032495e-05,
96
+ "loss": 0.056,
97
+ "step": 110
98
+ },
99
+ {
100
+ "epoch": 0.21631365479945922,
101
+ "grad_norm": 0.2994736135005951,
102
+ "learning_rate": 4.458483754512636e-05,
103
+ "loss": 0.0697,
104
+ "step": 120
105
+ },
106
+ {
107
+ "epoch": 0.23433979269941416,
108
+ "grad_norm": 0.5501758456230164,
109
+ "learning_rate": 4.413357400722022e-05,
110
+ "loss": 0.0518,
111
+ "step": 130
112
+ },
113
+ {
114
+ "epoch": 0.25236593059936907,
115
+ "grad_norm": 6.304769992828369,
116
+ "learning_rate": 4.368231046931408e-05,
117
+ "loss": 0.0583,
118
+ "step": 140
119
+ },
120
+ {
121
+ "epoch": 0.270392068499324,
122
+ "grad_norm": 0.6018352508544922,
123
+ "learning_rate": 4.3231046931407945e-05,
124
+ "loss": 0.0455,
125
+ "step": 150
126
+ },
127
+ {
128
+ "epoch": 0.28841820639927895,
129
+ "grad_norm": 5.612204551696777,
130
+ "learning_rate": 4.277978339350181e-05,
131
+ "loss": 0.0415,
132
+ "step": 160
133
+ },
134
+ {
135
+ "epoch": 0.3064443442992339,
136
+ "grad_norm": 8.740653991699219,
137
+ "learning_rate": 4.232851985559567e-05,
138
+ "loss": 0.0435,
139
+ "step": 170
140
+ },
141
+ {
142
+ "epoch": 0.32447048219918884,
143
+ "grad_norm": 0.3686186373233795,
144
+ "learning_rate": 4.187725631768953e-05,
145
+ "loss": 0.0629,
146
+ "step": 180
147
+ },
148
+ {
149
+ "epoch": 0.3424966200991438,
150
+ "grad_norm": 0.5254208445549011,
151
+ "learning_rate": 4.1425992779783394e-05,
152
+ "loss": 0.0544,
153
+ "step": 190
154
+ },
155
+ {
156
+ "epoch": 0.3605227579990987,
157
+ "grad_norm": 6.638070583343506,
158
+ "learning_rate": 4.0974729241877256e-05,
159
+ "loss": 0.0833,
160
+ "step": 200
161
+ },
162
+ {
163
+ "epoch": 0.3785488958990536,
164
+ "grad_norm": 7.016050815582275,
165
+ "learning_rate": 4.052346570397112e-05,
166
+ "loss": 0.0641,
167
+ "step": 210
168
+ },
169
+ {
170
+ "epoch": 0.39657503379900855,
171
+ "grad_norm": 0.49279534816741943,
172
+ "learning_rate": 4.007220216606498e-05,
173
+ "loss": 0.031,
174
+ "step": 220
175
+ },
176
+ {
177
+ "epoch": 0.4146011716989635,
178
+ "grad_norm": 2.8472509384155273,
179
+ "learning_rate": 3.962093862815885e-05,
180
+ "loss": 0.081,
181
+ "step": 230
182
+ },
183
+ {
184
+ "epoch": 0.43262730959891843,
185
+ "grad_norm": 3.4148313999176025,
186
+ "learning_rate": 3.916967509025271e-05,
187
+ "loss": 0.0579,
188
+ "step": 240
189
+ },
190
+ {
191
+ "epoch": 0.45065344749887337,
192
+ "grad_norm": 0.18055088818073273,
193
+ "learning_rate": 3.8718411552346575e-05,
194
+ "loss": 0.0238,
195
+ "step": 250
196
+ },
197
+ {
198
+ "epoch": 0.4686795853988283,
199
+ "grad_norm": 0.2409915030002594,
200
+ "learning_rate": 3.826714801444044e-05,
201
+ "loss": 0.0179,
202
+ "step": 260
203
+ },
204
+ {
205
+ "epoch": 0.48670572329878325,
206
+ "grad_norm": 10.743939399719238,
207
+ "learning_rate": 3.78158844765343e-05,
208
+ "loss": 0.0578,
209
+ "step": 270
210
+ },
211
+ {
212
+ "epoch": 0.5047318611987381,
213
+ "grad_norm": 20.40981674194336,
214
+ "learning_rate": 3.7364620938628155e-05,
215
+ "loss": 0.0667,
216
+ "step": 280
217
+ },
218
+ {
219
+ "epoch": 0.5227579990986931,
220
+ "grad_norm": 0.8752731680870056,
221
+ "learning_rate": 3.6913357400722025e-05,
222
+ "loss": 0.0312,
223
+ "step": 290
224
+ },
225
+ {
226
+ "epoch": 0.540784136998648,
227
+ "grad_norm": 4.817637920379639,
228
+ "learning_rate": 3.646209386281589e-05,
229
+ "loss": 0.0517,
230
+ "step": 300
231
+ },
232
+ {
233
+ "epoch": 0.558810274898603,
234
+ "grad_norm": 6.971864700317383,
235
+ "learning_rate": 3.601083032490975e-05,
236
+ "loss": 0.0344,
237
+ "step": 310
238
+ },
239
+ {
240
+ "epoch": 0.5768364127985579,
241
+ "grad_norm": 0.742476224899292,
242
+ "learning_rate": 3.555956678700361e-05,
243
+ "loss": 0.0561,
244
+ "step": 320
245
+ },
246
+ {
247
+ "epoch": 0.5948625506985128,
248
+ "grad_norm": 0.7727406024932861,
249
+ "learning_rate": 3.5108303249097474e-05,
250
+ "loss": 0.0617,
251
+ "step": 330
252
+ },
253
+ {
254
+ "epoch": 0.6128886885984678,
255
+ "grad_norm": 0.6510533094406128,
256
+ "learning_rate": 3.4657039711191336e-05,
257
+ "loss": 0.0551,
258
+ "step": 340
259
+ },
260
+ {
261
+ "epoch": 0.6309148264984227,
262
+ "grad_norm": 4.241283893585205,
263
+ "learning_rate": 3.42057761732852e-05,
264
+ "loss": 0.026,
265
+ "step": 350
266
+ },
267
+ {
268
+ "epoch": 0.6489409643983777,
269
+ "grad_norm": 0.18860740959644318,
270
+ "learning_rate": 3.375451263537907e-05,
271
+ "loss": 0.0367,
272
+ "step": 360
273
+ },
274
+ {
275
+ "epoch": 0.6669671022983326,
276
+ "grad_norm": 0.13881151378154755,
277
+ "learning_rate": 3.330324909747293e-05,
278
+ "loss": 0.0434,
279
+ "step": 370
280
+ },
281
+ {
282
+ "epoch": 0.6849932401982876,
283
+ "grad_norm": 0.45671743154525757,
284
+ "learning_rate": 3.2851985559566786e-05,
285
+ "loss": 0.049,
286
+ "step": 380
287
+ },
288
+ {
289
+ "epoch": 0.7030193780982424,
290
+ "grad_norm": 0.5025729537010193,
291
+ "learning_rate": 3.240072202166065e-05,
292
+ "loss": 0.0266,
293
+ "step": 390
294
+ },
295
+ {
296
+ "epoch": 0.7210455159981974,
297
+ "grad_norm": 0.3955366611480713,
298
+ "learning_rate": 3.194945848375451e-05,
299
+ "loss": 0.0264,
300
+ "step": 400
301
+ },
302
+ {
303
+ "epoch": 0.7390716538981523,
304
+ "grad_norm": 11.814117431640625,
305
+ "learning_rate": 3.149819494584837e-05,
306
+ "loss": 0.0495,
307
+ "step": 410
308
+ },
309
+ {
310
+ "epoch": 0.7570977917981072,
311
+ "grad_norm": 10.024807929992676,
312
+ "learning_rate": 3.104693140794224e-05,
313
+ "loss": 0.0262,
314
+ "step": 420
315
+ },
316
+ {
317
+ "epoch": 0.7751239296980622,
318
+ "grad_norm": 0.23220416903495789,
319
+ "learning_rate": 3.0595667870036104e-05,
320
+ "loss": 0.0826,
321
+ "step": 430
322
+ },
323
+ {
324
+ "epoch": 0.7931500675980171,
325
+ "grad_norm": 1.059012532234192,
326
+ "learning_rate": 3.0144404332129967e-05,
327
+ "loss": 0.0323,
328
+ "step": 440
329
+ },
330
+ {
331
+ "epoch": 0.8111762054979721,
332
+ "grad_norm": 5.872659683227539,
333
+ "learning_rate": 2.969314079422383e-05,
334
+ "loss": 0.0616,
335
+ "step": 450
336
+ },
337
+ {
338
+ "epoch": 0.829202343397927,
339
+ "grad_norm": 0.3091956377029419,
340
+ "learning_rate": 2.924187725631769e-05,
341
+ "loss": 0.0439,
342
+ "step": 460
343
+ },
344
+ {
345
+ "epoch": 0.847228481297882,
346
+ "grad_norm": 0.7430059909820557,
347
+ "learning_rate": 2.879061371841155e-05,
348
+ "loss": 0.0758,
349
+ "step": 470
350
+ },
351
+ {
352
+ "epoch": 0.8652546191978369,
353
+ "grad_norm": 0.5902143120765686,
354
+ "learning_rate": 2.8339350180505413e-05,
355
+ "loss": 0.0448,
356
+ "step": 480
357
+ },
358
+ {
359
+ "epoch": 0.8832807570977917,
360
+ "grad_norm": 5.876249313354492,
361
+ "learning_rate": 2.7888086642599282e-05,
362
+ "loss": 0.0356,
363
+ "step": 490
364
+ },
365
+ {
366
+ "epoch": 0.9013068949977467,
367
+ "grad_norm": 3.5948667526245117,
368
+ "learning_rate": 2.7436823104693144e-05,
369
+ "loss": 0.053,
370
+ "step": 500
371
+ },
372
+ {
373
+ "epoch": 0.9193330328977016,
374
+ "grad_norm": 9.958101272583008,
375
+ "learning_rate": 2.6985559566787007e-05,
376
+ "loss": 0.0234,
377
+ "step": 510
378
+ },
379
+ {
380
+ "epoch": 0.9373591707976566,
381
+ "grad_norm": 0.7074028253555298,
382
+ "learning_rate": 2.6534296028880866e-05,
383
+ "loss": 0.052,
384
+ "step": 520
385
+ },
386
+ {
387
+ "epoch": 0.9553853086976115,
388
+ "grad_norm": 0.37822696566581726,
389
+ "learning_rate": 2.6083032490974728e-05,
390
+ "loss": 0.0517,
391
+ "step": 530
392
+ },
393
+ {
394
+ "epoch": 0.9734114465975665,
395
+ "grad_norm": 0.5355275869369507,
396
+ "learning_rate": 2.563176895306859e-05,
397
+ "loss": 0.0327,
398
+ "step": 540
399
+ },
400
+ {
401
+ "epoch": 0.9914375844975214,
402
+ "grad_norm": 0.4926086366176605,
403
+ "learning_rate": 2.518050541516246e-05,
404
+ "loss": 0.042,
405
+ "step": 550
406
+ },
407
+ {
408
+ "epoch": 1.0,
409
+ "eval_f1": 0.0,
410
+ "eval_loss": 0.04069153591990471,
411
+ "eval_precision": 0.0,
412
+ "eval_recall": 0.0,
413
+ "eval_runtime": 142.0447,
414
+ "eval_samples_per_second": 32.898,
415
+ "eval_steps_per_second": 4.118,
416
+ "step": 555
417
+ },
418
+ {
419
+ "epoch": 1.0090130689499774,
420
+ "grad_norm": 0.7699225544929504,
421
+ "learning_rate": 2.472924187725632e-05,
422
+ "loss": 0.0093,
423
+ "step": 560
424
+ },
425
+ {
426
+ "epoch": 1.0270392068499323,
427
+ "grad_norm": 2.445814609527588,
428
+ "learning_rate": 2.427797833935018e-05,
429
+ "loss": 0.0348,
430
+ "step": 570
431
+ },
432
+ {
433
+ "epoch": 1.0450653447498874,
434
+ "grad_norm": 0.38662949204444885,
435
+ "learning_rate": 2.3826714801444043e-05,
436
+ "loss": 0.0165,
437
+ "step": 580
438
+ },
439
+ {
440
+ "epoch": 1.0630914826498423,
441
+ "grad_norm": 0.1346297264099121,
442
+ "learning_rate": 2.3375451263537906e-05,
443
+ "loss": 0.0759,
444
+ "step": 590
445
+ },
446
+ {
447
+ "epoch": 1.0811176205497972,
448
+ "grad_norm": 12.862547874450684,
449
+ "learning_rate": 2.292418772563177e-05,
450
+ "loss": 0.0296,
451
+ "step": 600
452
+ },
453
+ {
454
+ "epoch": 1.099143758449752,
455
+ "grad_norm": 3.661592960357666,
456
+ "learning_rate": 2.2472924187725634e-05,
457
+ "loss": 0.0182,
458
+ "step": 610
459
+ },
460
+ {
461
+ "epoch": 1.117169896349707,
462
+ "grad_norm": 4.039327144622803,
463
+ "learning_rate": 2.2021660649819496e-05,
464
+ "loss": 0.0484,
465
+ "step": 620
466
+ },
467
+ {
468
+ "epoch": 1.135196034249662,
469
+ "grad_norm": 7.463378429412842,
470
+ "learning_rate": 2.157039711191336e-05,
471
+ "loss": 0.051,
472
+ "step": 630
473
+ },
474
+ {
475
+ "epoch": 1.153222172149617,
476
+ "grad_norm": 4.8609724044799805,
477
+ "learning_rate": 2.111913357400722e-05,
478
+ "loss": 0.0641,
479
+ "step": 640
480
+ },
481
+ {
482
+ "epoch": 1.1712483100495719,
483
+ "grad_norm": 10.486072540283203,
484
+ "learning_rate": 2.0667870036101083e-05,
485
+ "loss": 0.0626,
486
+ "step": 650
487
+ },
488
+ {
489
+ "epoch": 1.1892744479495267,
490
+ "grad_norm": 0.24464073777198792,
491
+ "learning_rate": 2.0216606498194946e-05,
492
+ "loss": 0.0193,
493
+ "step": 660
494
+ },
495
+ {
496
+ "epoch": 1.2073005858494819,
497
+ "grad_norm": 3.8871383666992188,
498
+ "learning_rate": 1.976534296028881e-05,
499
+ "loss": 0.0303,
500
+ "step": 670
501
+ },
502
+ {
503
+ "epoch": 1.2253267237494367,
504
+ "grad_norm": 0.4887860119342804,
505
+ "learning_rate": 1.9314079422382674e-05,
506
+ "loss": 0.0447,
507
+ "step": 680
508
+ },
509
+ {
510
+ "epoch": 1.2433528616493916,
511
+ "grad_norm": 0.1695612519979477,
512
+ "learning_rate": 1.8862815884476533e-05,
513
+ "loss": 0.0327,
514
+ "step": 690
515
+ },
516
+ {
517
+ "epoch": 1.2613789995493465,
518
+ "grad_norm": 9.335827827453613,
519
+ "learning_rate": 1.84115523465704e-05,
520
+ "loss": 0.0186,
521
+ "step": 700
522
+ },
523
+ {
524
+ "epoch": 1.2794051374493014,
525
+ "grad_norm": 0.8768898248672485,
526
+ "learning_rate": 1.796028880866426e-05,
527
+ "loss": 0.0582,
528
+ "step": 710
529
+ },
530
+ {
531
+ "epoch": 1.2974312753492563,
532
+ "grad_norm": 0.18744328618049622,
533
+ "learning_rate": 1.7509025270758123e-05,
534
+ "loss": 0.0019,
535
+ "step": 720
536
+ },
537
+ {
538
+ "epoch": 1.3154574132492114,
539
+ "grad_norm": 6.990164756774902,
540
+ "learning_rate": 1.7057761732851986e-05,
541
+ "loss": 0.0601,
542
+ "step": 730
543
+ },
544
+ {
545
+ "epoch": 1.3334835511491663,
546
+ "grad_norm": 7.397508144378662,
547
+ "learning_rate": 1.6606498194945848e-05,
548
+ "loss": 0.0841,
549
+ "step": 740
550
+ },
551
+ {
552
+ "epoch": 1.3515096890491212,
553
+ "grad_norm": 0.5567547082901001,
554
+ "learning_rate": 1.615523465703971e-05,
555
+ "loss": 0.0356,
556
+ "step": 750
557
+ },
558
+ {
559
+ "epoch": 1.3695358269490763,
560
+ "grad_norm": 0.42192819714546204,
561
+ "learning_rate": 1.5703971119133576e-05,
562
+ "loss": 0.0094,
563
+ "step": 760
564
+ },
565
+ {
566
+ "epoch": 1.3875619648490312,
567
+ "grad_norm": 5.491224765777588,
568
+ "learning_rate": 1.5252707581227438e-05,
569
+ "loss": 0.0319,
570
+ "step": 770
571
+ },
572
+ {
573
+ "epoch": 1.405588102748986,
574
+ "grad_norm": 0.10327385365962982,
575
+ "learning_rate": 1.4801444043321299e-05,
576
+ "loss": 0.0282,
577
+ "step": 780
578
+ },
579
+ {
580
+ "epoch": 1.423614240648941,
581
+ "grad_norm": 0.2903471887111664,
582
+ "learning_rate": 1.4350180505415165e-05,
583
+ "loss": 0.0444,
584
+ "step": 790
585
+ },
586
+ {
587
+ "epoch": 1.4416403785488958,
588
+ "grad_norm": 5.516872882843018,
589
+ "learning_rate": 1.3898916967509026e-05,
590
+ "loss": 0.068,
591
+ "step": 800
592
+ },
593
+ {
594
+ "epoch": 1.4596665164488507,
595
+ "grad_norm": 0.15897555649280548,
596
+ "learning_rate": 1.3447653429602888e-05,
597
+ "loss": 0.0011,
598
+ "step": 810
599
+ },
600
+ {
601
+ "epoch": 1.4776926543488058,
602
+ "grad_norm": 9.496695518493652,
603
+ "learning_rate": 1.299638989169675e-05,
604
+ "loss": 0.0535,
605
+ "step": 820
606
+ },
607
+ {
608
+ "epoch": 1.4957187922487607,
609
+ "grad_norm": 0.23458808660507202,
610
+ "learning_rate": 1.2545126353790614e-05,
611
+ "loss": 0.0353,
612
+ "step": 830
613
+ },
614
+ {
615
+ "epoch": 1.5137449301487156,
616
+ "grad_norm": 6.693735599517822,
617
+ "learning_rate": 1.2093862815884477e-05,
618
+ "loss": 0.0235,
619
+ "step": 840
620
+ },
621
+ {
622
+ "epoch": 1.5317710680486707,
623
+ "grad_norm": 7.055171966552734,
624
+ "learning_rate": 1.164259927797834e-05,
625
+ "loss": 0.0315,
626
+ "step": 850
627
+ },
628
+ {
629
+ "epoch": 1.5497972059486256,
630
+ "grad_norm": 0.1647280752658844,
631
+ "learning_rate": 1.1191335740072201e-05,
632
+ "loss": 0.041,
633
+ "step": 860
634
+ },
635
+ {
636
+ "epoch": 1.5678233438485805,
637
+ "grad_norm": 31.603675842285156,
638
+ "learning_rate": 1.0740072202166066e-05,
639
+ "loss": 0.0637,
640
+ "step": 870
641
+ },
642
+ {
643
+ "epoch": 1.5858494817485354,
644
+ "grad_norm": 5.558256149291992,
645
+ "learning_rate": 1.028880866425993e-05,
646
+ "loss": 0.0534,
647
+ "step": 880
648
+ },
649
+ {
650
+ "epoch": 1.6038756196484902,
651
+ "grad_norm": 4.783972263336182,
652
+ "learning_rate": 9.83754512635379e-06,
653
+ "loss": 0.0382,
654
+ "step": 890
655
+ },
656
+ {
657
+ "epoch": 1.6219017575484451,
658
+ "grad_norm": 5.161861419677734,
659
+ "learning_rate": 9.386281588447654e-06,
660
+ "loss": 0.0454,
661
+ "step": 900
662
+ },
663
+ {
664
+ "epoch": 1.6399278954484002,
665
+ "grad_norm": 5.796766757965088,
666
+ "learning_rate": 8.935018050541517e-06,
667
+ "loss": 0.0221,
668
+ "step": 910
669
+ },
670
+ {
671
+ "epoch": 1.6579540333483551,
672
+ "grad_norm": 0.6868887543678284,
673
+ "learning_rate": 8.483754512635379e-06,
674
+ "loss": 0.028,
675
+ "step": 920
676
+ },
677
+ {
678
+ "epoch": 1.67598017124831,
679
+ "grad_norm": 0.27316200733184814,
680
+ "learning_rate": 8.032490974729243e-06,
681
+ "loss": 0.0234,
682
+ "step": 930
683
+ },
684
+ {
685
+ "epoch": 1.694006309148265,
686
+ "grad_norm": 0.4934704303741455,
687
+ "learning_rate": 7.581227436823105e-06,
688
+ "loss": 0.0171,
689
+ "step": 940
690
+ },
691
+ {
692
+ "epoch": 1.71203244704822,
693
+ "grad_norm": 0.4003954231739044,
694
+ "learning_rate": 7.129963898916968e-06,
695
+ "loss": 0.0159,
696
+ "step": 950
697
+ },
698
+ {
699
+ "epoch": 1.7300585849481749,
700
+ "grad_norm": 0.2824706733226776,
701
+ "learning_rate": 6.678700361010831e-06,
702
+ "loss": 0.0853,
703
+ "step": 960
704
+ },
705
+ {
706
+ "epoch": 1.7480847228481298,
707
+ "grad_norm": 0.14133542776107788,
708
+ "learning_rate": 6.227436823104693e-06,
709
+ "loss": 0.032,
710
+ "step": 970
711
+ },
712
+ {
713
+ "epoch": 1.7661108607480847,
714
+ "grad_norm": 0.30818939208984375,
715
+ "learning_rate": 5.776173285198557e-06,
716
+ "loss": 0.0258,
717
+ "step": 980
718
+ },
719
+ {
720
+ "epoch": 1.7841369986480395,
721
+ "grad_norm": 0.1608804315328598,
722
+ "learning_rate": 5.324909747292419e-06,
723
+ "loss": 0.0598,
724
+ "step": 990
725
+ },
726
+ {
727
+ "epoch": 1.8021631365479944,
728
+ "grad_norm": 33.14769744873047,
729
+ "learning_rate": 4.873646209386281e-06,
730
+ "loss": 0.0409,
731
+ "step": 1000
732
+ },
733
+ {
734
+ "epoch": 1.8201892744479495,
735
+ "grad_norm": 0.9574671983718872,
736
+ "learning_rate": 4.422382671480145e-06,
737
+ "loss": 0.0245,
738
+ "step": 1010
739
+ },
740
+ {
741
+ "epoch": 1.8382154123479044,
742
+ "grad_norm": 0.41375967860221863,
743
+ "learning_rate": 3.971119133574008e-06,
744
+ "loss": 0.0377,
745
+ "step": 1020
746
+ },
747
+ {
748
+ "epoch": 1.8562415502478595,
749
+ "grad_norm": 3.852210760116577,
750
+ "learning_rate": 3.51985559566787e-06,
751
+ "loss": 0.0358,
752
+ "step": 1030
753
+ },
754
+ {
755
+ "epoch": 1.8742676881478144,
756
+ "grad_norm": 7.285823822021484,
757
+ "learning_rate": 3.068592057761733e-06,
758
+ "loss": 0.0617,
759
+ "step": 1040
760
+ },
761
+ {
762
+ "epoch": 1.8922938260477693,
763
+ "grad_norm": 0.22026002407073975,
764
+ "learning_rate": 2.6173285198555957e-06,
765
+ "loss": 0.0423,
766
+ "step": 1050
767
+ },
768
+ {
769
+ "epoch": 1.9103199639477242,
770
+ "grad_norm": 0.30483102798461914,
771
+ "learning_rate": 2.1660649819494585e-06,
772
+ "loss": 0.0083,
773
+ "step": 1060
774
+ },
775
+ {
776
+ "epoch": 1.928346101847679,
777
+ "grad_norm": 2.322383165359497,
778
+ "learning_rate": 1.7148014440433215e-06,
779
+ "loss": 0.0112,
780
+ "step": 1070
781
+ },
782
+ {
783
+ "epoch": 1.946372239747634,
784
+ "grad_norm": 0.2074129730463028,
785
+ "learning_rate": 1.263537906137184e-06,
786
+ "loss": 0.0314,
787
+ "step": 1080
788
+ },
789
+ {
790
+ "epoch": 1.9643983776475888,
791
+ "grad_norm": 0.543554425239563,
792
+ "learning_rate": 8.122743682310469e-07,
793
+ "loss": 0.0244,
794
+ "step": 1090
795
+ },
796
+ {
797
+ "epoch": 1.982424515547544,
798
+ "grad_norm": 0.8434537053108215,
799
+ "learning_rate": 3.610108303249098e-07,
800
+ "loss": 0.0453,
801
+ "step": 1100
802
+ }
803
+ ],
804
+ "logging_steps": 10,
805
+ "max_steps": 1108,
806
+ "num_input_tokens_seen": 0,
807
+ "num_train_epochs": 2,
808
+ "save_steps": 500,
809
+ "stateful_callbacks": {
810
+ "TrainerControl": {
811
+ "args": {
812
+ "should_epoch_stop": false,
813
+ "should_evaluate": false,
814
+ "should_log": false,
815
+ "should_save": true,
816
+ "should_training_stop": true
817
+ },
818
+ "attributes": {}
819
+ }
820
+ },
821
+ "total_flos": 7.68611172962304e+16,
822
+ "train_batch_size": 8,
823
+ "trial_name": null,
824
+ "trial_params": null
825
+ }
checkpoint-1108/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97afd7a398103d448d50fc55ef878deb079a18d40d6865b60599e2b7a529812d
3
+ size 5432