anoaky commited on
Commit
213a1e4
1 Parent(s): 75ae630

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:d1d22c769651c52bfd5817957bb4f60c583517f050a059ffc0eef4475fbc2251
3
+ size 2954947648
checkpoint-1108/trainer_state.json ADDED
@@ -0,0 +1,848 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 0.1456953642384106,
14
+ "eval_fn": 11,
15
+ "eval_fp": 4117,
16
+ "eval_loss": 0.7862394452095032,
17
+ "eval_precision": 0.07876482434549116,
18
+ "eval_recall": 0.9696969696969697,
19
+ "eval_runtime": 48.4563,
20
+ "eval_samples_per_second": 96.437,
21
+ "eval_steps_per_second": 6.047,
22
+ "eval_tn": 193,
23
+ "eval_tp": 352,
24
+ "step": 0
25
+ },
26
+ {
27
+ "epoch": 0.018026137899954935,
28
+ "grad_norm": 5.399981498718262,
29
+ "learning_rate": 4.9548736462093865e-05,
30
+ "loss": 0.5179,
31
+ "step": 10
32
+ },
33
+ {
34
+ "epoch": 0.03605227579990987,
35
+ "grad_norm": 1.6319377422332764,
36
+ "learning_rate": 4.909747292418773e-05,
37
+ "loss": 0.2969,
38
+ "step": 20
39
+ },
40
+ {
41
+ "epoch": 0.054078413699864804,
42
+ "grad_norm": 1.8271431922912598,
43
+ "learning_rate": 4.864620938628159e-05,
44
+ "loss": 0.3075,
45
+ "step": 30
46
+ },
47
+ {
48
+ "epoch": 0.07210455159981974,
49
+ "grad_norm": 2.5912694931030273,
50
+ "learning_rate": 4.819494584837546e-05,
51
+ "loss": 0.3096,
52
+ "step": 40
53
+ },
54
+ {
55
+ "epoch": 0.09013068949977468,
56
+ "grad_norm": 3.3950061798095703,
57
+ "learning_rate": 4.7743682310469314e-05,
58
+ "loss": 0.2585,
59
+ "step": 50
60
+ },
61
+ {
62
+ "epoch": 0.10815682739972961,
63
+ "grad_norm": 1.6037261486053467,
64
+ "learning_rate": 4.7292418772563177e-05,
65
+ "loss": 0.2461,
66
+ "step": 60
67
+ },
68
+ {
69
+ "epoch": 0.12618296529968454,
70
+ "grad_norm": 3.6747143268585205,
71
+ "learning_rate": 4.684115523465704e-05,
72
+ "loss": 0.2803,
73
+ "step": 70
74
+ },
75
+ {
76
+ "epoch": 0.14420910319963948,
77
+ "grad_norm": 6.330898284912109,
78
+ "learning_rate": 4.63898916967509e-05,
79
+ "loss": 0.2735,
80
+ "step": 80
81
+ },
82
+ {
83
+ "epoch": 0.16223524109959442,
84
+ "grad_norm": 4.724136829376221,
85
+ "learning_rate": 4.5938628158844764e-05,
86
+ "loss": 0.2551,
87
+ "step": 90
88
+ },
89
+ {
90
+ "epoch": 0.18026137899954936,
91
+ "grad_norm": 3.852949619293213,
92
+ "learning_rate": 4.548736462093863e-05,
93
+ "loss": 0.2286,
94
+ "step": 100
95
+ },
96
+ {
97
+ "epoch": 0.19828751689950427,
98
+ "grad_norm": 10.743219375610352,
99
+ "learning_rate": 4.5036101083032495e-05,
100
+ "loss": 0.2511,
101
+ "step": 110
102
+ },
103
+ {
104
+ "epoch": 0.21631365479945922,
105
+ "grad_norm": 24.62398910522461,
106
+ "learning_rate": 4.458483754512636e-05,
107
+ "loss": 0.1756,
108
+ "step": 120
109
+ },
110
+ {
111
+ "epoch": 0.23433979269941416,
112
+ "grad_norm": 6.864269256591797,
113
+ "learning_rate": 4.413357400722022e-05,
114
+ "loss": 0.1947,
115
+ "step": 130
116
+ },
117
+ {
118
+ "epoch": 0.25236593059936907,
119
+ "grad_norm": 11.810012817382812,
120
+ "learning_rate": 4.368231046931408e-05,
121
+ "loss": 0.1976,
122
+ "step": 140
123
+ },
124
+ {
125
+ "epoch": 0.270392068499324,
126
+ "grad_norm": 4.794726848602295,
127
+ "learning_rate": 4.3231046931407945e-05,
128
+ "loss": 0.2081,
129
+ "step": 150
130
+ },
131
+ {
132
+ "epoch": 0.28841820639927895,
133
+ "grad_norm": 7.1115875244140625,
134
+ "learning_rate": 4.277978339350181e-05,
135
+ "loss": 0.2055,
136
+ "step": 160
137
+ },
138
+ {
139
+ "epoch": 0.3064443442992339,
140
+ "grad_norm": 6.199843883514404,
141
+ "learning_rate": 4.232851985559567e-05,
142
+ "loss": 0.1764,
143
+ "step": 170
144
+ },
145
+ {
146
+ "epoch": 0.32447048219918884,
147
+ "grad_norm": 5.7516679763793945,
148
+ "learning_rate": 4.187725631768953e-05,
149
+ "loss": 0.185,
150
+ "step": 180
151
+ },
152
+ {
153
+ "epoch": 0.3424966200991438,
154
+ "grad_norm": 17.204120635986328,
155
+ "learning_rate": 4.1425992779783394e-05,
156
+ "loss": 0.175,
157
+ "step": 190
158
+ },
159
+ {
160
+ "epoch": 0.3605227579990987,
161
+ "grad_norm": 6.416110038757324,
162
+ "learning_rate": 4.0974729241877256e-05,
163
+ "loss": 0.1197,
164
+ "step": 200
165
+ },
166
+ {
167
+ "epoch": 0.3785488958990536,
168
+ "grad_norm": 8.525187492370605,
169
+ "learning_rate": 4.052346570397112e-05,
170
+ "loss": 0.1968,
171
+ "step": 210
172
+ },
173
+ {
174
+ "epoch": 0.39657503379900855,
175
+ "grad_norm": 42.04795455932617,
176
+ "learning_rate": 4.007220216606498e-05,
177
+ "loss": 0.1718,
178
+ "step": 220
179
+ },
180
+ {
181
+ "epoch": 0.4146011716989635,
182
+ "grad_norm": 4.502872467041016,
183
+ "learning_rate": 3.962093862815885e-05,
184
+ "loss": 0.1442,
185
+ "step": 230
186
+ },
187
+ {
188
+ "epoch": 0.43262730959891843,
189
+ "grad_norm": 3.215198278427124,
190
+ "learning_rate": 3.916967509025271e-05,
191
+ "loss": 0.1675,
192
+ "step": 240
193
+ },
194
+ {
195
+ "epoch": 0.45065344749887337,
196
+ "grad_norm": 6.314309120178223,
197
+ "learning_rate": 3.8718411552346575e-05,
198
+ "loss": 0.1492,
199
+ "step": 250
200
+ },
201
+ {
202
+ "epoch": 0.4686795853988283,
203
+ "grad_norm": 8.839702606201172,
204
+ "learning_rate": 3.826714801444044e-05,
205
+ "loss": 0.1399,
206
+ "step": 260
207
+ },
208
+ {
209
+ "epoch": 0.48670572329878325,
210
+ "grad_norm": 183.9261016845703,
211
+ "learning_rate": 3.78158844765343e-05,
212
+ "loss": 0.1801,
213
+ "step": 270
214
+ },
215
+ {
216
+ "epoch": 0.5047318611987381,
217
+ "grad_norm": 5.393545150756836,
218
+ "learning_rate": 3.7364620938628155e-05,
219
+ "loss": 0.1353,
220
+ "step": 280
221
+ },
222
+ {
223
+ "epoch": 0.5227579990986931,
224
+ "grad_norm": 3.587885856628418,
225
+ "learning_rate": 3.6913357400722025e-05,
226
+ "loss": 0.1428,
227
+ "step": 290
228
+ },
229
+ {
230
+ "epoch": 0.540784136998648,
231
+ "grad_norm": 9.456450462341309,
232
+ "learning_rate": 3.646209386281589e-05,
233
+ "loss": 0.1758,
234
+ "step": 300
235
+ },
236
+ {
237
+ "epoch": 0.558810274898603,
238
+ "grad_norm": 4.051924705505371,
239
+ "learning_rate": 3.601083032490975e-05,
240
+ "loss": 0.1874,
241
+ "step": 310
242
+ },
243
+ {
244
+ "epoch": 0.5768364127985579,
245
+ "grad_norm": 6.4390363693237305,
246
+ "learning_rate": 3.555956678700361e-05,
247
+ "loss": 0.1472,
248
+ "step": 320
249
+ },
250
+ {
251
+ "epoch": 0.5948625506985128,
252
+ "grad_norm": 3.86102557182312,
253
+ "learning_rate": 3.5108303249097474e-05,
254
+ "loss": 0.1659,
255
+ "step": 330
256
+ },
257
+ {
258
+ "epoch": 0.6128886885984678,
259
+ "grad_norm": 6.6616644859313965,
260
+ "learning_rate": 3.4657039711191336e-05,
261
+ "loss": 0.191,
262
+ "step": 340
263
+ },
264
+ {
265
+ "epoch": 0.6309148264984227,
266
+ "grad_norm": 6.867959976196289,
267
+ "learning_rate": 3.42057761732852e-05,
268
+ "loss": 0.1343,
269
+ "step": 350
270
+ },
271
+ {
272
+ "epoch": 0.6489409643983777,
273
+ "grad_norm": 8.724226951599121,
274
+ "learning_rate": 3.375451263537907e-05,
275
+ "loss": 0.145,
276
+ "step": 360
277
+ },
278
+ {
279
+ "epoch": 0.6669671022983326,
280
+ "grad_norm": 3.5500874519348145,
281
+ "learning_rate": 3.330324909747293e-05,
282
+ "loss": 0.1153,
283
+ "step": 370
284
+ },
285
+ {
286
+ "epoch": 0.6849932401982876,
287
+ "grad_norm": 9.760895729064941,
288
+ "learning_rate": 3.2851985559566786e-05,
289
+ "loss": 0.1362,
290
+ "step": 380
291
+ },
292
+ {
293
+ "epoch": 0.7030193780982424,
294
+ "grad_norm": 3.9465765953063965,
295
+ "learning_rate": 3.240072202166065e-05,
296
+ "loss": 0.1281,
297
+ "step": 390
298
+ },
299
+ {
300
+ "epoch": 0.7210455159981974,
301
+ "grad_norm": 3.495382070541382,
302
+ "learning_rate": 3.194945848375451e-05,
303
+ "loss": 0.1792,
304
+ "step": 400
305
+ },
306
+ {
307
+ "epoch": 0.7390716538981523,
308
+ "grad_norm": 2.8898510932922363,
309
+ "learning_rate": 3.149819494584837e-05,
310
+ "loss": 0.1085,
311
+ "step": 410
312
+ },
313
+ {
314
+ "epoch": 0.7570977917981072,
315
+ "grad_norm": 3.3261382579803467,
316
+ "learning_rate": 3.104693140794224e-05,
317
+ "loss": 0.203,
318
+ "step": 420
319
+ },
320
+ {
321
+ "epoch": 0.7751239296980622,
322
+ "grad_norm": 3.417875051498413,
323
+ "learning_rate": 3.0595667870036104e-05,
324
+ "loss": 0.1491,
325
+ "step": 430
326
+ },
327
+ {
328
+ "epoch": 0.7931500675980171,
329
+ "grad_norm": 7.868630409240723,
330
+ "learning_rate": 3.0144404332129967e-05,
331
+ "loss": 0.1777,
332
+ "step": 440
333
+ },
334
+ {
335
+ "epoch": 0.8111762054979721,
336
+ "grad_norm": 2.16607403755188,
337
+ "learning_rate": 2.969314079422383e-05,
338
+ "loss": 0.1406,
339
+ "step": 450
340
+ },
341
+ {
342
+ "epoch": 0.829202343397927,
343
+ "grad_norm": 5.121480464935303,
344
+ "learning_rate": 2.924187725631769e-05,
345
+ "loss": 0.1387,
346
+ "step": 460
347
+ },
348
+ {
349
+ "epoch": 0.847228481297882,
350
+ "grad_norm": 6.610152244567871,
351
+ "learning_rate": 2.879061371841155e-05,
352
+ "loss": 0.1158,
353
+ "step": 470
354
+ },
355
+ {
356
+ "epoch": 0.8652546191978369,
357
+ "grad_norm": 5.962329387664795,
358
+ "learning_rate": 2.8339350180505413e-05,
359
+ "loss": 0.127,
360
+ "step": 480
361
+ },
362
+ {
363
+ "epoch": 0.8832807570977917,
364
+ "grad_norm": 5.11064338684082,
365
+ "learning_rate": 2.7888086642599282e-05,
366
+ "loss": 0.1232,
367
+ "step": 490
368
+ },
369
+ {
370
+ "epoch": 0.9013068949977467,
371
+ "grad_norm": 5.128263473510742,
372
+ "learning_rate": 2.7436823104693144e-05,
373
+ "loss": 0.1084,
374
+ "step": 500
375
+ },
376
+ {
377
+ "epoch": 0.9193330328977016,
378
+ "grad_norm": 8.249767303466797,
379
+ "learning_rate": 2.6985559566787007e-05,
380
+ "loss": 0.1453,
381
+ "step": 510
382
+ },
383
+ {
384
+ "epoch": 0.9373591707976566,
385
+ "grad_norm": 14.850152969360352,
386
+ "learning_rate": 2.6534296028880866e-05,
387
+ "loss": 0.1102,
388
+ "step": 520
389
+ },
390
+ {
391
+ "epoch": 0.9553853086976115,
392
+ "grad_norm": 4.934476852416992,
393
+ "learning_rate": 2.6083032490974728e-05,
394
+ "loss": 0.1493,
395
+ "step": 530
396
+ },
397
+ {
398
+ "epoch": 0.9734114465975665,
399
+ "grad_norm": 6.984730243682861,
400
+ "learning_rate": 2.563176895306859e-05,
401
+ "loss": 0.1344,
402
+ "step": 540
403
+ },
404
+ {
405
+ "epoch": 0.9914375844975214,
406
+ "grad_norm": 4.595370292663574,
407
+ "learning_rate": 2.518050541516246e-05,
408
+ "loss": 0.1246,
409
+ "step": 550
410
+ },
411
+ {
412
+ "epoch": 1.0,
413
+ "eval_f1": 0.7272727272727273,
414
+ "eval_fn": 95,
415
+ "eval_fp": 106,
416
+ "eval_loss": 0.11311017721891403,
417
+ "eval_precision": 0.7165775401069518,
418
+ "eval_recall": 0.7382920110192838,
419
+ "eval_runtime": 56.8141,
420
+ "eval_samples_per_second": 82.251,
421
+ "eval_steps_per_second": 5.157,
422
+ "eval_tn": 4204,
423
+ "eval_tp": 268,
424
+ "step": 555
425
+ },
426
+ {
427
+ "epoch": 1.0090130689499774,
428
+ "grad_norm": 3.8490426540374756,
429
+ "learning_rate": 2.472924187725632e-05,
430
+ "loss": 0.131,
431
+ "step": 560
432
+ },
433
+ {
434
+ "epoch": 1.0270392068499323,
435
+ "grad_norm": 8.608445167541504,
436
+ "learning_rate": 2.427797833935018e-05,
437
+ "loss": 0.0894,
438
+ "step": 570
439
+ },
440
+ {
441
+ "epoch": 1.0450653447498874,
442
+ "grad_norm": 7.926303863525391,
443
+ "learning_rate": 2.3826714801444043e-05,
444
+ "loss": 0.1462,
445
+ "step": 580
446
+ },
447
+ {
448
+ "epoch": 1.0630914826498423,
449
+ "grad_norm": 5.411970138549805,
450
+ "learning_rate": 2.3375451263537906e-05,
451
+ "loss": 0.1556,
452
+ "step": 590
453
+ },
454
+ {
455
+ "epoch": 1.0811176205497972,
456
+ "grad_norm": 6.707615375518799,
457
+ "learning_rate": 2.292418772563177e-05,
458
+ "loss": 0.0999,
459
+ "step": 600
460
+ },
461
+ {
462
+ "epoch": 1.099143758449752,
463
+ "grad_norm": 5.215950012207031,
464
+ "learning_rate": 2.2472924187725634e-05,
465
+ "loss": 0.0854,
466
+ "step": 610
467
+ },
468
+ {
469
+ "epoch": 1.117169896349707,
470
+ "grad_norm": 1.6493467092514038,
471
+ "learning_rate": 2.2021660649819496e-05,
472
+ "loss": 0.0806,
473
+ "step": 620
474
+ },
475
+ {
476
+ "epoch": 1.135196034249662,
477
+ "grad_norm": 8.559683799743652,
478
+ "learning_rate": 2.157039711191336e-05,
479
+ "loss": 0.1422,
480
+ "step": 630
481
+ },
482
+ {
483
+ "epoch": 1.153222172149617,
484
+ "grad_norm": 3.0874404907226562,
485
+ "learning_rate": 2.111913357400722e-05,
486
+ "loss": 0.1436,
487
+ "step": 640
488
+ },
489
+ {
490
+ "epoch": 1.1712483100495719,
491
+ "grad_norm": 13.975804328918457,
492
+ "learning_rate": 2.0667870036101083e-05,
493
+ "loss": 0.1294,
494
+ "step": 650
495
+ },
496
+ {
497
+ "epoch": 1.1892744479495267,
498
+ "grad_norm": 52.61249923706055,
499
+ "learning_rate": 2.0216606498194946e-05,
500
+ "loss": 0.127,
501
+ "step": 660
502
+ },
503
+ {
504
+ "epoch": 1.2073005858494819,
505
+ "grad_norm": 4.427943706512451,
506
+ "learning_rate": 1.976534296028881e-05,
507
+ "loss": 0.1101,
508
+ "step": 670
509
+ },
510
+ {
511
+ "epoch": 1.2253267237494367,
512
+ "grad_norm": 3.466806411743164,
513
+ "learning_rate": 1.9314079422382674e-05,
514
+ "loss": 0.0733,
515
+ "step": 680
516
+ },
517
+ {
518
+ "epoch": 1.2433528616493916,
519
+ "grad_norm": 7.252458572387695,
520
+ "learning_rate": 1.8862815884476533e-05,
521
+ "loss": 0.1129,
522
+ "step": 690
523
+ },
524
+ {
525
+ "epoch": 1.2613789995493465,
526
+ "grad_norm": 7.521577835083008,
527
+ "learning_rate": 1.84115523465704e-05,
528
+ "loss": 0.0894,
529
+ "step": 700
530
+ },
531
+ {
532
+ "epoch": 1.2794051374493014,
533
+ "grad_norm": 2.96766996383667,
534
+ "learning_rate": 1.796028880866426e-05,
535
+ "loss": 0.0828,
536
+ "step": 710
537
+ },
538
+ {
539
+ "epoch": 1.2974312753492563,
540
+ "grad_norm": 3.5544350147247314,
541
+ "learning_rate": 1.7509025270758123e-05,
542
+ "loss": 0.1205,
543
+ "step": 720
544
+ },
545
+ {
546
+ "epoch": 1.3154574132492114,
547
+ "grad_norm": 2.323742389678955,
548
+ "learning_rate": 1.7057761732851986e-05,
549
+ "loss": 0.1114,
550
+ "step": 730
551
+ },
552
+ {
553
+ "epoch": 1.3334835511491663,
554
+ "grad_norm": 4.532501220703125,
555
+ "learning_rate": 1.6606498194945848e-05,
556
+ "loss": 0.1242,
557
+ "step": 740
558
+ },
559
+ {
560
+ "epoch": 1.3515096890491212,
561
+ "grad_norm": 4.962907314300537,
562
+ "learning_rate": 1.615523465703971e-05,
563
+ "loss": 0.1222,
564
+ "step": 750
565
+ },
566
+ {
567
+ "epoch": 1.3695358269490763,
568
+ "grad_norm": 2.745126724243164,
569
+ "learning_rate": 1.5703971119133576e-05,
570
+ "loss": 0.1089,
571
+ "step": 760
572
+ },
573
+ {
574
+ "epoch": 1.3875619648490312,
575
+ "grad_norm": 1.2508844137191772,
576
+ "learning_rate": 1.5252707581227438e-05,
577
+ "loss": 0.086,
578
+ "step": 770
579
+ },
580
+ {
581
+ "epoch": 1.405588102748986,
582
+ "grad_norm": 9.282389640808105,
583
+ "learning_rate": 1.4801444043321299e-05,
584
+ "loss": 0.1142,
585
+ "step": 780
586
+ },
587
+ {
588
+ "epoch": 1.423614240648941,
589
+ "grad_norm": 3.1791019439697266,
590
+ "learning_rate": 1.4350180505415165e-05,
591
+ "loss": 0.1237,
592
+ "step": 790
593
+ },
594
+ {
595
+ "epoch": 1.4416403785488958,
596
+ "grad_norm": 6.121169567108154,
597
+ "learning_rate": 1.3898916967509026e-05,
598
+ "loss": 0.1346,
599
+ "step": 800
600
+ },
601
+ {
602
+ "epoch": 1.4596665164488507,
603
+ "grad_norm": 19.356491088867188,
604
+ "learning_rate": 1.3447653429602888e-05,
605
+ "loss": 0.0937,
606
+ "step": 810
607
+ },
608
+ {
609
+ "epoch": 1.4776926543488058,
610
+ "grad_norm": 6.563851356506348,
611
+ "learning_rate": 1.299638989169675e-05,
612
+ "loss": 0.1136,
613
+ "step": 820
614
+ },
615
+ {
616
+ "epoch": 1.4957187922487607,
617
+ "grad_norm": 5.794551372528076,
618
+ "learning_rate": 1.2545126353790614e-05,
619
+ "loss": 0.119,
620
+ "step": 830
621
+ },
622
+ {
623
+ "epoch": 1.5137449301487156,
624
+ "grad_norm": 2.885023832321167,
625
+ "learning_rate": 1.2093862815884477e-05,
626
+ "loss": 0.1073,
627
+ "step": 840
628
+ },
629
+ {
630
+ "epoch": 1.5317710680486707,
631
+ "grad_norm": 3.519407272338867,
632
+ "learning_rate": 1.164259927797834e-05,
633
+ "loss": 0.122,
634
+ "step": 850
635
+ },
636
+ {
637
+ "epoch": 1.5497972059486256,
638
+ "grad_norm": 3.4222164154052734,
639
+ "learning_rate": 1.1191335740072201e-05,
640
+ "loss": 0.1002,
641
+ "step": 860
642
+ },
643
+ {
644
+ "epoch": 1.5678233438485805,
645
+ "grad_norm": 8.442163467407227,
646
+ "learning_rate": 1.0740072202166066e-05,
647
+ "loss": 0.1301,
648
+ "step": 870
649
+ },
650
+ {
651
+ "epoch": 1.5858494817485354,
652
+ "grad_norm": 5.987030982971191,
653
+ "learning_rate": 1.028880866425993e-05,
654
+ "loss": 0.1197,
655
+ "step": 880
656
+ },
657
+ {
658
+ "epoch": 1.6038756196484902,
659
+ "grad_norm": 3.641470432281494,
660
+ "learning_rate": 9.83754512635379e-06,
661
+ "loss": 0.094,
662
+ "step": 890
663
+ },
664
+ {
665
+ "epoch": 1.6219017575484451,
666
+ "grad_norm": 4.365111827850342,
667
+ "learning_rate": 9.386281588447654e-06,
668
+ "loss": 0.1309,
669
+ "step": 900
670
+ },
671
+ {
672
+ "epoch": 1.6399278954484002,
673
+ "grad_norm": 6.229074478149414,
674
+ "learning_rate": 8.935018050541517e-06,
675
+ "loss": 0.1002,
676
+ "step": 910
677
+ },
678
+ {
679
+ "epoch": 1.6579540333483551,
680
+ "grad_norm": 8.32435417175293,
681
+ "learning_rate": 8.483754512635379e-06,
682
+ "loss": 0.1155,
683
+ "step": 920
684
+ },
685
+ {
686
+ "epoch": 1.67598017124831,
687
+ "grad_norm": 2.6587719917297363,
688
+ "learning_rate": 8.032490974729243e-06,
689
+ "loss": 0.1047,
690
+ "step": 930
691
+ },
692
+ {
693
+ "epoch": 1.694006309148265,
694
+ "grad_norm": 31.095243453979492,
695
+ "learning_rate": 7.581227436823105e-06,
696
+ "loss": 0.0957,
697
+ "step": 940
698
+ },
699
+ {
700
+ "epoch": 1.71203244704822,
701
+ "grad_norm": 148.5176544189453,
702
+ "learning_rate": 7.129963898916968e-06,
703
+ "loss": 0.1138,
704
+ "step": 950
705
+ },
706
+ {
707
+ "epoch": 1.7300585849481749,
708
+ "grad_norm": 4.901301383972168,
709
+ "learning_rate": 6.678700361010831e-06,
710
+ "loss": 0.0939,
711
+ "step": 960
712
+ },
713
+ {
714
+ "epoch": 1.7480847228481298,
715
+ "grad_norm": 4.844261646270752,
716
+ "learning_rate": 6.227436823104693e-06,
717
+ "loss": 0.0905,
718
+ "step": 970
719
+ },
720
+ {
721
+ "epoch": 1.7661108607480847,
722
+ "grad_norm": 6.100379943847656,
723
+ "learning_rate": 5.776173285198557e-06,
724
+ "loss": 0.1169,
725
+ "step": 980
726
+ },
727
+ {
728
+ "epoch": 1.7841369986480395,
729
+ "grad_norm": 9.385725021362305,
730
+ "learning_rate": 5.324909747292419e-06,
731
+ "loss": 0.0843,
732
+ "step": 990
733
+ },
734
+ {
735
+ "epoch": 1.8021631365479944,
736
+ "grad_norm": 8.386088371276855,
737
+ "learning_rate": 4.873646209386281e-06,
738
+ "loss": 0.1044,
739
+ "step": 1000
740
+ },
741
+ {
742
+ "epoch": 1.8201892744479495,
743
+ "grad_norm": 13.418025970458984,
744
+ "learning_rate": 4.422382671480145e-06,
745
+ "loss": 0.1351,
746
+ "step": 1010
747
+ },
748
+ {
749
+ "epoch": 1.8382154123479044,
750
+ "grad_norm": 13.804243087768555,
751
+ "learning_rate": 3.971119133574008e-06,
752
+ "loss": 0.0919,
753
+ "step": 1020
754
+ },
755
+ {
756
+ "epoch": 1.8562415502478595,
757
+ "grad_norm": 7.5179924964904785,
758
+ "learning_rate": 3.51985559566787e-06,
759
+ "loss": 0.1305,
760
+ "step": 1030
761
+ },
762
+ {
763
+ "epoch": 1.8742676881478144,
764
+ "grad_norm": 4.629436016082764,
765
+ "learning_rate": 3.068592057761733e-06,
766
+ "loss": 0.1278,
767
+ "step": 1040
768
+ },
769
+ {
770
+ "epoch": 1.8922938260477693,
771
+ "grad_norm": 5.896274566650391,
772
+ "learning_rate": 2.6173285198555957e-06,
773
+ "loss": 0.1211,
774
+ "step": 1050
775
+ },
776
+ {
777
+ "epoch": 1.9103199639477242,
778
+ "grad_norm": 7.580524921417236,
779
+ "learning_rate": 2.1660649819494585e-06,
780
+ "loss": 0.0993,
781
+ "step": 1060
782
+ },
783
+ {
784
+ "epoch": 1.928346101847679,
785
+ "grad_norm": 12.92769718170166,
786
+ "learning_rate": 1.7148014440433215e-06,
787
+ "loss": 0.1557,
788
+ "step": 1070
789
+ },
790
+ {
791
+ "epoch": 1.946372239747634,
792
+ "grad_norm": 12.059624671936035,
793
+ "learning_rate": 1.263537906137184e-06,
794
+ "loss": 0.0929,
795
+ "step": 1080
796
+ },
797
+ {
798
+ "epoch": 1.9643983776475888,
799
+ "grad_norm": 4.273173809051514,
800
+ "learning_rate": 8.122743682310469e-07,
801
+ "loss": 0.1053,
802
+ "step": 1090
803
+ },
804
+ {
805
+ "epoch": 1.982424515547544,
806
+ "grad_norm": 7.6728596687316895,
807
+ "learning_rate": 3.610108303249098e-07,
808
+ "loss": 0.0997,
809
+ "step": 1100
810
+ },
811
+ {
812
+ "epoch": 1.996845425867508,
813
+ "eval_f1": 0.7513513513513513,
814
+ "eval_fn": 85,
815
+ "eval_fp": 99,
816
+ "eval_loss": 0.10865739732980728,
817
+ "eval_precision": 0.7374005305039788,
818
+ "eval_recall": 0.7658402203856749,
819
+ "eval_runtime": 59.9339,
820
+ "eval_samples_per_second": 77.969,
821
+ "eval_steps_per_second": 4.889,
822
+ "eval_tn": 4211,
823
+ "eval_tp": 278,
824
+ "step": 1108
825
+ }
826
+ ],
827
+ "logging_steps": 10,
828
+ "max_steps": 1108,
829
+ "num_input_tokens_seen": 0,
830
+ "num_train_epochs": 2,
831
+ "save_steps": 500,
832
+ "stateful_callbacks": {
833
+ "TrainerControl": {
834
+ "args": {
835
+ "should_epoch_stop": false,
836
+ "should_evaluate": false,
837
+ "should_log": false,
838
+ "should_save": true,
839
+ "should_training_stop": true
840
+ },
841
+ "attributes": {}
842
+ }
843
+ },
844
+ "total_flos": 7.68611172962304e+16,
845
+ "train_batch_size": 16,
846
+ "trial_name": null,
847
+ "trial_params": null
848
+ }
checkpoint-1108/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76059b5f2bd4f1247a190d00bab4a2d1780e8431aa1970375f0ea6a2bd29d309
3
+ size 5368