terry69 commited on
Commit
b1143e5
·
verified ·
1 Parent(s): fc3e016

Model save

Browse files
Files changed (4) hide show
  1. README.md +69 -0
  2. all_results.json +9 -0
  3. train_results.json +9 -0
  4. trainer_state.json +701 -0
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: llama3
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ base_model: meta-llama/Meta-Llama-3-8B
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: downstream_0.1p_seed42_level1_style
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # downstream_0.1p_seed42_level1_style
20
+
21
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.9802
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0002
43
+ - train_batch_size: 8
44
+ - eval_batch_size: 1
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 4
48
+ - gradient_accumulation_steps: 4
49
+ - total_train_batch_size: 128
50
+ - total_eval_batch_size: 4
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: cosine
53
+ - lr_scheduler_warmup_ratio: 0.1
54
+ - num_epochs: 1
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss |
59
+ |:-------------:|:------:|:----:|:---------------:|
60
+ | 1.1043 | 0.9984 | 460 | 0.9802 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - PEFT 0.11.1
66
+ - Transformers 4.43.4
67
+ - Pytorch 2.3.1+cu121
68
+ - Datasets 2.19.1
69
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9983722192078134,
3
+ "total_flos": 1.4599365322604544e+16,
4
+ "train_loss": 1.1265571905219036,
5
+ "train_runtime": 17446.1728,
6
+ "train_samples": 100000,
7
+ "train_samples_per_second": 3.379,
8
+ "train_steps_per_second": 0.026
9
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9983722192078134,
3
+ "total_flos": 1.4599365322604544e+16,
4
+ "train_loss": 1.1265571905219036,
5
+ "train_runtime": 17446.1728,
6
+ "train_samples": 100000,
7
+ "train_samples_per_second": 3.379,
8
+ "train_steps_per_second": 0.026
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,701 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9983722192078134,
5
+ "eval_steps": 500,
6
+ "global_step": 460,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.002170374389582203,
13
+ "grad_norm": 1.069300659689101,
14
+ "learning_rate": 4.347826086956522e-06,
15
+ "loss": 1.414,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.010851871947911014,
20
+ "grad_norm": 1.0869761401892497,
21
+ "learning_rate": 2.173913043478261e-05,
22
+ "loss": 1.3963,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.02170374389582203,
27
+ "grad_norm": 0.2444703280571688,
28
+ "learning_rate": 4.347826086956522e-05,
29
+ "loss": 1.3471,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.032555615843733045,
34
+ "grad_norm": 0.21229931040050687,
35
+ "learning_rate": 6.521739130434783e-05,
36
+ "loss": 1.2864,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.04340748779164406,
41
+ "grad_norm": 0.1762825707559893,
42
+ "learning_rate": 8.695652173913044e-05,
43
+ "loss": 1.2743,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.05425935973955507,
48
+ "grad_norm": 0.16054804651124344,
49
+ "learning_rate": 0.00010869565217391305,
50
+ "loss": 1.2327,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.06511123168746609,
55
+ "grad_norm": 0.1039496796881345,
56
+ "learning_rate": 0.00013043478260869567,
57
+ "loss": 1.2126,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.0759631036353771,
62
+ "grad_norm": 0.09387480834859042,
63
+ "learning_rate": 0.00015217391304347827,
64
+ "loss": 1.1809,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.08681497558328811,
69
+ "grad_norm": 0.10738381024430724,
70
+ "learning_rate": 0.00017391304347826088,
71
+ "loss": 1.1958,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.09766684753119913,
76
+ "grad_norm": 0.07429080271089483,
77
+ "learning_rate": 0.0001956521739130435,
78
+ "loss": 1.1523,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.10851871947911014,
83
+ "grad_norm": 0.08592662653519074,
84
+ "learning_rate": 0.00019995393663024054,
85
+ "loss": 1.1594,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.11937059142702117,
90
+ "grad_norm": 0.09095607145429864,
91
+ "learning_rate": 0.00019976687691905393,
92
+ "loss": 1.1614,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.13022246337493218,
97
+ "grad_norm": 0.09659070176916736,
98
+ "learning_rate": 0.00019943621095573586,
99
+ "loss": 1.1479,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.1410743353228432,
104
+ "grad_norm": 0.08577984921023268,
105
+ "learning_rate": 0.0001989624147068713,
106
+ "loss": 1.1363,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.1519262072707542,
111
+ "grad_norm": 0.09436954934257955,
112
+ "learning_rate": 0.0001983461701633742,
113
+ "loss": 1.1407,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.16277807921866522,
118
+ "grad_norm": 0.07860010966590557,
119
+ "learning_rate": 0.00019758836435881746,
120
+ "loss": 1.1566,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.17362995116657623,
125
+ "grad_norm": 0.07271734829149594,
126
+ "learning_rate": 0.00019669008809262062,
127
+ "loss": 1.139,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.18448182311448724,
132
+ "grad_norm": 0.08286259363397955,
133
+ "learning_rate": 0.0001956526343599335,
134
+ "loss": 1.1428,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.19533369506239825,
139
+ "grad_norm": 0.07039798681596814,
140
+ "learning_rate": 0.00019447749649047542,
141
+ "loss": 1.128,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.20618556701030927,
146
+ "grad_norm": 0.07088917306719042,
147
+ "learning_rate": 0.00019316636599900946,
148
+ "loss": 1.1283,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.21703743895822028,
153
+ "grad_norm": 0.07293899658768029,
154
+ "learning_rate": 0.00019172113015054532,
155
+ "loss": 1.1229,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.22788931090613132,
160
+ "grad_norm": 0.07001959910529978,
161
+ "learning_rate": 0.00019014386924377582,
162
+ "loss": 1.1341,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.23874118285404233,
167
+ "grad_norm": 0.07897051254796171,
168
+ "learning_rate": 0.00018843685361665723,
169
+ "loss": 1.1367,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.24959305480195335,
174
+ "grad_norm": 0.07980670912072682,
175
+ "learning_rate": 0.00018660254037844388,
176
+ "loss": 1.1198,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.26044492674986436,
181
+ "grad_norm": 0.07327615168352668,
182
+ "learning_rate": 0.00018464356987288013,
183
+ "loss": 1.1226,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.27129679869777534,
188
+ "grad_norm": 0.07718729819586401,
189
+ "learning_rate": 0.00018256276187764197,
190
+ "loss": 1.1132,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.2821486706456864,
195
+ "grad_norm": 0.06955350279895621,
196
+ "learning_rate": 0.00018036311154549784,
197
+ "loss": 1.1143,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.29300054259359737,
202
+ "grad_norm": 0.07312017765794143,
203
+ "learning_rate": 0.00017804778509303138,
204
+ "loss": 1.1055,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.3038524145415084,
209
+ "grad_norm": 0.06917446214768698,
210
+ "learning_rate": 0.00017562011524313185,
211
+ "loss": 1.1036,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.31470428648941945,
216
+ "grad_norm": 0.07416178127350893,
217
+ "learning_rate": 0.00017308359642781242,
218
+ "loss": 1.1105,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.32555615843733043,
223
+ "grad_norm": 0.07262036822274683,
224
+ "learning_rate": 0.00017044187975826124,
225
+ "loss": 1.1219,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.3364080303852415,
230
+ "grad_norm": 0.07372734639616982,
231
+ "learning_rate": 0.0001676987677693659,
232
+ "loss": 1.1089,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.34725990233315246,
237
+ "grad_norm": 0.07467616325072211,
238
+ "learning_rate": 0.0001648582089462756,
239
+ "loss": 1.1242,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.3581117742810635,
244
+ "grad_norm": 0.07693038799730868,
245
+ "learning_rate": 0.0001619242920408802,
246
+ "loss": 1.1035,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.3689636462289745,
251
+ "grad_norm": 0.07824471460501765,
252
+ "learning_rate": 0.00015890124018638638,
253
+ "loss": 1.1215,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.3798155181768855,
258
+ "grad_norm": 0.07888988968904209,
259
+ "learning_rate": 0.00015579340481846336,
260
+ "loss": 1.1255,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.3906673901247965,
265
+ "grad_norm": 0.0793792296368889,
266
+ "learning_rate": 0.00015260525941170712,
267
+ "loss": 1.1172,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.40151926207270755,
272
+ "grad_norm": 0.0787057550750279,
273
+ "learning_rate": 0.00014934139304044033,
274
+ "loss": 1.1081,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.41237113402061853,
279
+ "grad_norm": 0.07367605233129934,
280
+ "learning_rate": 0.00014600650377311522,
281
+ "loss": 1.1213,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.4232230059685296,
286
+ "grad_norm": 0.07155404224760639,
287
+ "learning_rate": 0.00014260539190982886,
288
+ "loss": 1.1023,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.43407487791644056,
293
+ "grad_norm": 0.07019186009135339,
294
+ "learning_rate": 0.00013914295307268396,
295
+ "loss": 1.1104,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.4449267498643516,
300
+ "grad_norm": 0.07069181937055558,
301
+ "learning_rate": 0.00013562417115894172,
302
+ "loss": 1.1078,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.45577862181226264,
307
+ "grad_norm": 0.07290667661940653,
308
+ "learning_rate": 0.00013205411116710972,
309
+ "loss": 1.0888,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.4666304937601736,
314
+ "grad_norm": 0.07161072440717098,
315
+ "learning_rate": 0.0001284379119062912,
316
+ "loss": 1.1235,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.47748236570808467,
321
+ "grad_norm": 0.07060414902470838,
322
+ "learning_rate": 0.00012478077859929,
323
+ "loss": 1.1172,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.48833423765599565,
328
+ "grad_norm": 0.07615333425469985,
329
+ "learning_rate": 0.00012108797539011847,
330
+ "loss": 1.1098,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.4991861096039067,
335
+ "grad_norm": 0.07182595107605282,
336
+ "learning_rate": 0.00011736481776669306,
337
+ "loss": 1.1132,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.5100379815518177,
342
+ "grad_norm": 0.0744050275020255,
343
+ "learning_rate": 0.00011361666490962468,
344
+ "loss": 1.1058,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.5208898534997287,
349
+ "grad_norm": 0.06916253752221385,
350
+ "learning_rate": 0.00010984891197811687,
351
+ "loss": 1.1073,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.5317417254476398,
356
+ "grad_norm": 0.07255483864108674,
357
+ "learning_rate": 0.00010606698234407586,
358
+ "loss": 1.1159,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.5425935973955507,
363
+ "grad_norm": 0.0718643676883612,
364
+ "learning_rate": 0.00010227631978561056,
365
+ "loss": 1.1067,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.5534454693434617,
370
+ "grad_norm": 0.07709129734227686,
371
+ "learning_rate": 9.848238065115975e-05,
372
+ "loss": 1.1059,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.5642973412913728,
377
+ "grad_norm": 0.07119845087632411,
378
+ "learning_rate": 9.469062600552509e-05,
379
+ "loss": 1.1072,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.5751492132392838,
384
+ "grad_norm": 0.07116811249448173,
385
+ "learning_rate": 9.09065137691153e-05,
386
+ "loss": 1.0919,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.5860010851871947,
391
+ "grad_norm": 0.07826423399254606,
392
+ "learning_rate": 8.713549086171691e-05,
393
+ "loss": 1.0896,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.5968529571351058,
398
+ "grad_norm": 0.06985572465886306,
399
+ "learning_rate": 8.33829853620986e-05,
400
+ "loss": 1.0826,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.6077048290830168,
405
+ "grad_norm": 0.07521227026642825,
406
+ "learning_rate": 7.965439869473664e-05,
407
+ "loss": 1.1015,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.6185567010309279,
412
+ "grad_norm": 0.07085025384725667,
413
+ "learning_rate": 7.595509785490617e-05,
414
+ "loss": 1.1082,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.6294085729788389,
419
+ "grad_norm": 0.07314117620701872,
420
+ "learning_rate": 7.229040768333115e-05,
421
+ "loss": 1.1191,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.6402604449267498,
426
+ "grad_norm": 0.07800277021405978,
427
+ "learning_rate": 6.866560320151179e-05,
428
+ "loss": 1.1217,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.6511123168746609,
433
+ "grad_norm": 0.07587494199122763,
434
+ "learning_rate": 6.508590201876317e-05,
435
+ "loss": 1.1187,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.6619641888225719,
440
+ "grad_norm": 0.07415053941958896,
441
+ "learning_rate": 6.155645682189351e-05,
442
+ "loss": 1.1124,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.672816060770483,
447
+ "grad_norm": 0.07124587825867733,
448
+ "learning_rate": 5.8082347958333625e-05,
449
+ "loss": 1.1062,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.6836679327183939,
454
+ "grad_norm": 0.07506903523794208,
455
+ "learning_rate": 5.466857612339229e-05,
456
+ "loss": 1.1227,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.6945198046663049,
461
+ "grad_norm": 0.07916490012178254,
462
+ "learning_rate": 5.1320055162165115e-05,
463
+ "loss": 1.1115,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.705371676614216,
468
+ "grad_norm": 0.07011216352533452,
469
+ "learning_rate": 4.804160499645667e-05,
470
+ "loss": 1.0982,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.716223548562127,
475
+ "grad_norm": 0.07092293644080847,
476
+ "learning_rate": 4.483794468689728e-05,
477
+ "loss": 1.0812,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.7270754205100379,
482
+ "grad_norm": 0.0718641240464739,
483
+ "learning_rate": 4.1713685640242165e-05,
484
+ "loss": 1.1113,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.737927292457949,
489
+ "grad_norm": 0.06926837229200397,
490
+ "learning_rate": 3.8673324971628357e-05,
491
+ "loss": 1.0968,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.74877916440586,
496
+ "grad_norm": 0.06788108886080438,
497
+ "learning_rate": 3.5721239031346066e-05,
498
+ "loss": 1.1169,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.759631036353771,
503
+ "grad_norm": 0.07009824066249028,
504
+ "learning_rate": 3.2861677105440336e-05,
505
+ "loss": 1.117,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.7704829083016821,
510
+ "grad_norm": 0.07291917614935897,
511
+ "learning_rate": 3.009875529921181e-05,
512
+ "loss": 1.1034,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.781334780249593,
517
+ "grad_norm": 0.07028920602015883,
518
+ "learning_rate": 2.7436450612420095e-05,
519
+ "loss": 1.0959,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.7921866521975041,
524
+ "grad_norm": 0.07447440376259351,
525
+ "learning_rate": 2.4878595214718236e-05,
526
+ "loss": 1.0976,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.8030385241454151,
531
+ "grad_norm": 0.06926375563549898,
532
+ "learning_rate": 2.242887092955801e-05,
533
+ "loss": 1.1142,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.8138903960933261,
538
+ "grad_norm": 0.06772391666168086,
539
+ "learning_rate": 2.0090803934506764e-05,
540
+ "loss": 1.1087,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.8247422680412371,
545
+ "grad_norm": 0.06992654447190966,
546
+ "learning_rate": 1.7867759685603114e-05,
547
+ "loss": 1.0827,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.8355941399891481,
552
+ "grad_norm": 0.07112396412052087,
553
+ "learning_rate": 1.5762938073058853e-05,
554
+ "loss": 1.1133,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.8464460119370592,
559
+ "grad_norm": 0.0690455035742087,
560
+ "learning_rate": 1.3779368815278647e-05,
561
+ "loss": 1.0997,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.8572978838849702,
566
+ "grad_norm": 0.06782571299859211,
567
+ "learning_rate": 1.1919907097828653e-05,
568
+ "loss": 1.0827,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.8681497558328811,
573
+ "grad_norm": 0.06820364734874568,
574
+ "learning_rate": 1.01872294636304e-05,
575
+ "loss": 1.1097,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.8790016277807922,
580
+ "grad_norm": 0.06714602286547451,
581
+ "learning_rate": 8.58382996029652e-06,
582
+ "loss": 1.1122,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.8898534997287032,
587
+ "grad_norm": 0.0684925645871222,
588
+ "learning_rate": 7.1120165501533e-06,
589
+ "loss": 1.0831,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.9007053716766142,
594
+ "grad_norm": 0.06878969210961007,
595
+ "learning_rate": 5.77390778811796e-06,
596
+ "loss": 1.0946,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.9115572436245253,
601
+ "grad_norm": 0.06870977921001499,
602
+ "learning_rate": 4.5714297722121106e-06,
603
+ "loss": 1.1182,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.9224091155724362,
608
+ "grad_norm": 0.0705672247176395,
609
+ "learning_rate": 3.5063133711014882e-06,
610
+ "loss": 1.1038,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.9332609875203473,
615
+ "grad_norm": 0.07691125665136789,
616
+ "learning_rate": 2.580091732652101e-06,
617
+ "loss": 1.1109,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.9441128594682583,
622
+ "grad_norm": 0.07114346398569787,
623
+ "learning_rate": 1.7940980770894122e-06,
624
+ "loss": 1.113,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.9549647314161693,
629
+ "grad_norm": 0.0706109763180513,
630
+ "learning_rate": 1.1494637779369766e-06,
631
+ "loss": 1.0971,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.9658166033640803,
636
+ "grad_norm": 0.06965109543647903,
637
+ "learning_rate": 6.471167334968886e-07,
638
+ "loss": 1.093,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.9766684753119913,
643
+ "grad_norm": 0.07100289385231237,
644
+ "learning_rate": 2.877800312160783e-07,
645
+ "loss": 1.109,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.9875203472599023,
650
+ "grad_norm": 0.07102279530958475,
651
+ "learning_rate": 7.197090686119623e-08,
652
+ "loss": 1.109,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.9983722192078134,
657
+ "grad_norm": 0.06941067748225418,
658
+ "learning_rate": 0.0,
659
+ "loss": 1.1043,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.9983722192078134,
664
+ "eval_loss": 0.9801881909370422,
665
+ "eval_runtime": 2.1363,
666
+ "eval_samples_per_second": 3.277,
667
+ "eval_steps_per_second": 0.936,
668
+ "step": 460
669
+ },
670
+ {
671
+ "epoch": 0.9983722192078134,
672
+ "step": 460,
673
+ "total_flos": 1.4599365322604544e+16,
674
+ "train_loss": 1.1265571905219036,
675
+ "train_runtime": 17446.1728,
676
+ "train_samples_per_second": 3.379,
677
+ "train_steps_per_second": 0.026
678
+ }
679
+ ],
680
+ "logging_steps": 5,
681
+ "max_steps": 460,
682
+ "num_input_tokens_seen": 0,
683
+ "num_train_epochs": 1,
684
+ "save_steps": 100,
685
+ "stateful_callbacks": {
686
+ "TrainerControl": {
687
+ "args": {
688
+ "should_epoch_stop": false,
689
+ "should_evaluate": false,
690
+ "should_log": false,
691
+ "should_save": true,
692
+ "should_training_stop": true
693
+ },
694
+ "attributes": {}
695
+ }
696
+ },
697
+ "total_flos": 1.4599365322604544e+16,
698
+ "train_batch_size": 8,
699
+ "trial_name": null,
700
+ "trial_params": null
701
+ }