geshijoker commited on
Commit
c5f9363
·
verified ·
1 Parent(s): 6f80bc6

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3
4
  base_model: meta-llama/Meta-Llama-3-8B-Instruct
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: lora_all
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # lora_all
17
 
18
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 1.7828
21
 
22
  ## Model description
23
 
 
4
  base_model: meta-llama/Meta-Llama-3-8B-Instruct
5
  tags:
6
  - llama-factory
7
+ - lora
8
  - generated_from_trainer
9
  model-index:
10
  - name: lora_all
 
16
 
17
  # lora_all
18
 
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the healthcaremagic dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 1.7827
22
 
23
  ## Model description
24
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.994667614646285,
3
+ "eval_loss": 1.782659888267517,
4
+ "eval_runtime": 74.5046,
5
+ "eval_samples_per_second": 134.22,
6
+ "eval_steps_per_second": 4.201,
7
+ "total_flos": 4.229776342129836e+18,
8
+ "train_loss": 1.8626822329427895,
9
+ "train_runtime": 5321.5336,
10
+ "train_samples_per_second": 50.737,
11
+ "train_steps_per_second": 0.198
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.994667614646285,
3
+ "eval_loss": 1.782659888267517,
4
+ "eval_runtime": 74.5046,
5
+ "eval_samples_per_second": 134.22,
6
+ "eval_steps_per_second": 4.201
7
+ }
runs/Dec29_00-33-03_64-181-222-101/events.out.tfevents.1735437830.64-181-222-101.10253.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f654033f3e0ec274d7e9786a21b66774f1c7da61cacb5c975fa6ebcff7a44b39
3
+ size 359
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.994667614646285,
3
+ "total_flos": 4.229776342129836e+18,
4
+ "train_loss": 1.8626822329427895,
5
+ "train_runtime": 5321.5336,
6
+ "train_samples_per_second": 50.737,
7
+ "train_steps_per_second": 0.198
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,785 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.994667614646285,
5
+ "eval_steps": 1000,
6
+ "global_step": 1053,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.028439388553146108,
13
+ "grad_norm": 2.7864015102386475,
14
+ "learning_rate": 9.433962264150944e-06,
15
+ "loss": 3.2455,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.056878777106292217,
20
+ "grad_norm": 0.8451234698295593,
21
+ "learning_rate": 1.8867924528301888e-05,
22
+ "loss": 2.8076,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.08531816565943832,
27
+ "grad_norm": 0.4598900079727173,
28
+ "learning_rate": 2.830188679245283e-05,
29
+ "loss": 2.5631,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.11375755421258443,
34
+ "grad_norm": 0.3573364317417145,
35
+ "learning_rate": 3.7735849056603776e-05,
36
+ "loss": 2.4723,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.14219694276573053,
41
+ "grad_norm": 0.3167368173599243,
42
+ "learning_rate": 4.716981132075472e-05,
43
+ "loss": 2.3806,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.17063633131887665,
48
+ "grad_norm": 0.3523406982421875,
49
+ "learning_rate": 4.999395511092461e-05,
50
+ "loss": 2.3182,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.19907571987202274,
55
+ "grad_norm": 0.3779957592487335,
56
+ "learning_rate": 4.996435452798774e-05,
57
+ "loss": 2.2473,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.22751510842516887,
62
+ "grad_norm": 0.4943179786205292,
63
+ "learning_rate": 4.991011714111481e-05,
64
+ "loss": 2.199,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.25595449697831496,
69
+ "grad_norm": 0.42343392968177795,
70
+ "learning_rate": 4.9831296476058484e-05,
71
+ "loss": 2.187,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.28439388553146105,
76
+ "grad_norm": 0.4268713593482971,
77
+ "learning_rate": 4.9727970319299044e-05,
78
+ "loss": 2.1194,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.3128332740846072,
83
+ "grad_norm": 0.48644664883613586,
84
+ "learning_rate": 4.9600240641278496e-05,
85
+ "loss": 2.1086,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.3412726626377533,
90
+ "grad_norm": 0.48994767665863037,
91
+ "learning_rate": 4.944823349576805e-05,
92
+ "loss": 2.0861,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.3697120511908994,
97
+ "grad_norm": 0.5868293642997742,
98
+ "learning_rate": 4.9272098895468277e-05,
99
+ "loss": 2.0809,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.3981514397440455,
104
+ "grad_norm": 0.5047050714492798,
105
+ "learning_rate": 4.907201066396469e-05,
106
+ "loss": 2.0588,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.42659082829719164,
111
+ "grad_norm": 0.4889000654220581,
112
+ "learning_rate": 4.8848166264184844e-05,
113
+ "loss": 2.019,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.45503021685033773,
118
+ "grad_norm": 0.5262014269828796,
119
+ "learning_rate": 4.860078660352625e-05,
120
+ "loss": 2.0294,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.4834696054034838,
125
+ "grad_norm": 0.4898081123828888,
126
+ "learning_rate": 4.8330115815847465e-05,
127
+ "loss": 1.9942,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.5119089939566299,
132
+ "grad_norm": 0.5089215040206909,
133
+ "learning_rate": 4.803642102053746e-05,
134
+ "loss": 1.9974,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.540348382509776,
139
+ "grad_norm": 0.5131760239601135,
140
+ "learning_rate": 4.7719992058901006e-05,
141
+ "loss": 1.9876,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.5687877710629221,
146
+ "grad_norm": 0.4940701723098755,
147
+ "learning_rate": 4.7381141208120296e-05,
148
+ "loss": 1.9438,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.5972271596160682,
153
+ "grad_norm": 0.5069682002067566,
154
+ "learning_rate": 4.702020287307509e-05,
155
+ "loss": 1.9668,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.6256665481692144,
160
+ "grad_norm": 0.5188295245170593,
161
+ "learning_rate": 4.663753325632548e-05,
162
+ "loss": 1.961,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.6541059367223605,
167
+ "grad_norm": 0.5184710025787354,
168
+ "learning_rate": 4.6233510006582914e-05,
169
+ "loss": 1.9327,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.6825453252755066,
174
+ "grad_norm": 0.5142987370491028,
175
+ "learning_rate": 4.580853184601659e-05,
176
+ "loss": 1.9415,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.7109847138286527,
181
+ "grad_norm": 0.5494813919067383,
182
+ "learning_rate": 4.536301817676274e-05,
183
+ "loss": 1.918,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.7394241023817988,
188
+ "grad_norm": 0.49111875891685486,
189
+ "learning_rate": 4.48974086670254e-05,
190
+ "loss": 1.9406,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.7678634909349449,
195
+ "grad_norm": 0.513344407081604,
196
+ "learning_rate": 4.4412162817176965e-05,
197
+ "loss": 1.9019,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.796302879488091,
202
+ "grad_norm": 0.5305178761482239,
203
+ "learning_rate": 4.39077595062868e-05,
204
+ "loss": 1.8887,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.8247422680412371,
209
+ "grad_norm": 0.5225845575332642,
210
+ "learning_rate": 4.33846965195254e-05,
211
+ "loss": 1.8942,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.8531816565943833,
216
+ "grad_norm": 0.5039849281311035,
217
+ "learning_rate": 4.2843490056910534e-05,
218
+ "loss": 1.8961,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.8816210451475294,
223
+ "grad_norm": 0.5015091896057129,
224
+ "learning_rate": 4.228467422388016e-05,
225
+ "loss": 1.9114,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.9100604337006755,
230
+ "grad_norm": 0.4691685140132904,
231
+ "learning_rate": 4.1708800504194827e-05,
232
+ "loss": 1.9162,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.9384998222538216,
237
+ "grad_norm": 0.48344337940216064,
238
+ "learning_rate": 4.1116437215689784e-05,
239
+ "loss": 1.8799,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.9669392108069677,
244
+ "grad_norm": 0.5020110011100769,
245
+ "learning_rate": 4.0508168949413906e-05,
246
+ "loss": 1.8565,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.9953785993601137,
251
+ "grad_norm": 0.514559268951416,
252
+ "learning_rate": 3.988459599270888e-05,
253
+ "loss": 1.9027,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.0238179879132598,
258
+ "grad_norm": 0.5024710893630981,
259
+ "learning_rate": 3.9246333736798095e-05,
260
+ "loss": 1.9138,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.052257376466406,
265
+ "grad_norm": 0.515842854976654,
266
+ "learning_rate": 3.859401206946982e-05,
267
+ "loss": 1.813,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.080696765019552,
272
+ "grad_norm": 0.5406286716461182,
273
+ "learning_rate": 3.792827475345393e-05,
274
+ "loss": 1.8395,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 1.1091361535726982,
279
+ "grad_norm": 0.5171347260475159,
280
+ "learning_rate": 3.724977879110591e-05,
281
+ "loss": 1.8314,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 1.1375755421258442,
286
+ "grad_norm": 0.5132310390472412,
287
+ "learning_rate": 3.6559193776024794e-05,
288
+ "loss": 1.8241,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 1.1660149306789904,
293
+ "grad_norm": 0.5178537368774414,
294
+ "learning_rate": 3.585720123224512e-05,
295
+ "loss": 1.8178,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 1.1944543192321366,
300
+ "grad_norm": 0.5596044063568115,
301
+ "learning_rate": 3.5144493941655e-05,
302
+ "loss": 1.8174,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.2228937077852826,
307
+ "grad_norm": 0.5178967714309692,
308
+ "learning_rate": 3.442177526030407e-05,
309
+ "loss": 1.7867,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 1.2513330963384286,
314
+ "grad_norm": 0.5135601758956909,
315
+ "learning_rate": 3.3689758424275926e-05,
316
+ "loss": 1.791,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 1.2797724848915748,
321
+ "grad_norm": 0.5257358551025391,
322
+ "learning_rate": 3.294916584581027e-05,
323
+ "loss": 1.8153,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 1.308211873444721,
328
+ "grad_norm": 0.5272982716560364,
329
+ "learning_rate": 3.220072840036923e-05,
330
+ "loss": 1.7952,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 1.336651261997867,
335
+ "grad_norm": 0.5305171012878418,
336
+ "learning_rate": 3.14451847053515e-05,
337
+ "loss": 1.7962,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 1.3650906505510132,
342
+ "grad_norm": 0.5360648036003113,
343
+ "learning_rate": 3.068328039116616e-05,
344
+ "loss": 1.8002,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 1.3935300391041592,
349
+ "grad_norm": 0.5958048105239868,
350
+ "learning_rate": 2.99157673653855e-05,
351
+ "loss": 1.8017,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 1.4219694276573054,
356
+ "grad_norm": 0.5489828586578369,
357
+ "learning_rate": 2.9143403070702997e-05,
358
+ "loss": 1.7931,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 1.4504088162104516,
363
+ "grad_norm": 0.5472132563591003,
364
+ "learning_rate": 2.8366949737428817e-05,
365
+ "loss": 1.8051,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 1.4788482047635976,
370
+ "grad_norm": 0.5372362732887268,
371
+ "learning_rate": 2.7587173631260566e-05,
372
+ "loss": 1.7962,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 1.5072875933167436,
377
+ "grad_norm": 0.5616655349731445,
378
+ "learning_rate": 2.6804844297071526e-05,
379
+ "loss": 1.7763,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 1.5357269818698898,
384
+ "grad_norm": 0.5398069620132446,
385
+ "learning_rate": 2.6020733799462754e-05,
386
+ "loss": 1.7808,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 1.564166370423036,
391
+ "grad_norm": 0.5471286773681641,
392
+ "learning_rate": 2.5235615960828605e-05,
393
+ "loss": 1.7836,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 1.5926057589761822,
398
+ "grad_norm": 0.5741537809371948,
399
+ "learning_rate": 2.4450265597687376e-05,
400
+ "loss": 1.8075,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 1.6210451475293282,
405
+ "grad_norm": 0.5587407946586609,
406
+ "learning_rate": 2.3665457756030988e-05,
407
+ "loss": 1.7669,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 1.6494845360824741,
412
+ "grad_norm": 0.5349502563476562,
413
+ "learning_rate": 2.2881966946448167e-05,
414
+ "loss": 1.7712,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 1.6779239246356203,
419
+ "grad_norm": 0.5548702478408813,
420
+ "learning_rate": 2.2100566379775967e-05,
421
+ "loss": 1.7669,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 1.7063633131887666,
426
+ "grad_norm": 0.5402134656906128,
427
+ "learning_rate": 2.1322027204034066e-05,
428
+ "loss": 1.7754,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 1.7348027017419125,
433
+ "grad_norm": 0.5542489290237427,
434
+ "learning_rate": 2.0547117743394744e-05,
435
+ "loss": 1.7959,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 1.7632420902950585,
440
+ "grad_norm": 0.6034718751907349,
441
+ "learning_rate": 1.9776602739939714e-05,
442
+ "loss": 1.7631,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 1.7916814788482047,
447
+ "grad_norm": 0.5417160987854004,
448
+ "learning_rate": 1.9011242598951962e-05,
449
+ "loss": 1.7681,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 1.820120867401351,
454
+ "grad_norm": 0.5751153826713562,
455
+ "learning_rate": 1.8251792638487596e-05,
456
+ "loss": 1.7717,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 1.8485602559544971,
461
+ "grad_norm": 0.517077624797821,
462
+ "learning_rate": 1.7499002343968098e-05,
463
+ "loss": 1.7571,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 1.8769996445076431,
468
+ "grad_norm": 0.5594078302383423,
469
+ "learning_rate": 1.675361462852868e-05,
470
+ "loss": 1.7528,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 1.905439033060789,
475
+ "grad_norm": 0.5302609801292419,
476
+ "learning_rate": 1.6016365099852735e-05,
477
+ "loss": 1.7454,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 1.9338784216139353,
482
+ "grad_norm": 0.5613446235656738,
483
+ "learning_rate": 1.528798133421585e-05,
484
+ "loss": 1.7555,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 1.9623178101670815,
489
+ "grad_norm": 0.5346989631652832,
490
+ "learning_rate": 1.4569182158455875e-05,
491
+ "loss": 1.7334,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 1.9907571987202275,
496
+ "grad_norm": 0.5548744201660156,
497
+ "learning_rate": 1.3860676940577594e-05,
498
+ "loss": 1.7764,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 2.0191965872733735,
503
+ "grad_norm": 0.6024349331855774,
504
+ "learning_rate": 1.3163164889692197e-05,
505
+ "loss": 1.7899,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 2.0476359758265197,
510
+ "grad_norm": 0.583003044128418,
511
+ "learning_rate": 1.2477334365982248e-05,
512
+ "loss": 1.7026,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 2.076075364379666,
517
+ "grad_norm": 0.572210967540741,
518
+ "learning_rate": 1.1803862201373342e-05,
519
+ "loss": 1.6817,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 2.104514752932812,
524
+ "grad_norm": 0.567309558391571,
525
+ "learning_rate": 1.1143413031582645e-05,
526
+ "loss": 1.702,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 2.132954141485958,
531
+ "grad_norm": 0.5702211856842041,
532
+ "learning_rate": 1.0496638640203774e-05,
533
+ "loss": 1.7001,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 2.161393530039104,
538
+ "grad_norm": 0.598107635974884,
539
+ "learning_rate": 9.864177315474968e-06,
540
+ "loss": 1.6904,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 2.1898329185922503,
545
+ "grad_norm": 0.5752361416816711,
546
+ "learning_rate": 9.246653220365778e-06,
547
+ "loss": 1.7187,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 2.2182723071453965,
552
+ "grad_norm": 0.5880258679389954,
553
+ "learning_rate": 8.644675776603476e-06,
554
+ "loss": 1.6973,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 2.2467116956985427,
559
+ "grad_norm": 0.5714329481124878,
560
+ "learning_rate": 8.058839063247447e-06,
561
+ "loss": 1.706,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 2.2751510842516884,
566
+ "grad_norm": 0.6069587469100952,
567
+ "learning_rate": 7.489721230404842e-06,
568
+ "loss": 1.7323,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 2.3035904728048346,
573
+ "grad_norm": 0.5789757966995239,
574
+ "learning_rate": 6.937883928666255e-06,
575
+ "loss": 1.7076,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 2.332029861357981,
580
+ "grad_norm": 0.6063619256019592,
581
+ "learning_rate": 6.403871754824373e-06,
582
+ "loss": 1.6819,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 2.360469249911127,
587
+ "grad_norm": 0.5955121517181396,
588
+ "learning_rate": 5.8882117144227115e-06,
589
+ "loss": 1.6991,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 2.3889086384642733,
594
+ "grad_norm": 0.5998035073280334,
595
+ "learning_rate": 5.391412701664744e-06,
596
+ "loss": 1.6747,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 2.417348027017419,
601
+ "grad_norm": 0.6063375473022461,
602
+ "learning_rate": 4.91396499719681e-06,
603
+ "loss": 1.7041,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 2.4457874155705652,
608
+ "grad_norm": 0.6151806712150574,
609
+ "learning_rate": 4.456339784260247e-06,
610
+ "loss": 1.6868,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 2.4742268041237114,
615
+ "grad_norm": 0.5840434432029724,
616
+ "learning_rate": 4.018988683690461e-06,
617
+ "loss": 1.6757,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 2.502666192676857,
622
+ "grad_norm": 0.5928480625152588,
623
+ "learning_rate": 3.6023433082216755e-06,
624
+ "loss": 1.702,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 2.5311055812300034,
629
+ "grad_norm": 0.5977081656455994,
630
+ "learning_rate": 3.2068148365372806e-06,
631
+ "loss": 1.7001,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 2.5595449697831496,
636
+ "grad_norm": 0.6291081309318542,
637
+ "learning_rate": 2.832793607486087e-06,
638
+ "loss": 1.7146,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 2.587984358336296,
643
+ "grad_norm": 0.6002483367919922,
644
+ "learning_rate": 2.4806487348650485e-06,
645
+ "loss": 1.6753,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 2.616423746889442,
650
+ "grad_norm": 0.5828536748886108,
651
+ "learning_rate": 2.150727743148473e-06,
652
+ "loss": 1.6863,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 2.6448631354425878,
657
+ "grad_norm": 0.6066195964813232,
658
+ "learning_rate": 1.8433562245233349e-06,
659
+ "loss": 1.658,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 2.673302523995734,
664
+ "grad_norm": 0.6164836287498474,
665
+ "learning_rate": 1.5588375175691117e-06,
666
+ "loss": 1.6957,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 2.70174191254888,
671
+ "grad_norm": 0.6047420501708984,
672
+ "learning_rate": 1.2974524078991995e-06,
673
+ "loss": 1.677,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 2.7301813011020264,
678
+ "grad_norm": 0.5752055048942566,
679
+ "learning_rate": 1.0594588510594445e-06,
680
+ "loss": 1.6802,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 2.7586206896551726,
685
+ "grad_norm": 0.5959452986717224,
686
+ "learning_rate": 8.450917179571305e-07,
687
+ "loss": 1.6897,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 2.7870600782083184,
692
+ "grad_norm": 0.5909944176673889,
693
+ "learning_rate": 6.545625630717783e-07,
694
+ "loss": 1.6916,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 2.8154994667614646,
699
+ "grad_norm": 0.601028323173523,
700
+ "learning_rate": 4.880594156763896e-07,
701
+ "loss": 1.7078,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 2.8439388553146108,
706
+ "grad_norm": 0.5776922702789307,
707
+ "learning_rate": 3.4574659427528133e-07,
708
+ "loss": 1.6961,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 2.8439388553146108,
713
+ "eval_loss": 1.7827636003494263,
714
+ "eval_runtime": 74.8387,
715
+ "eval_samples_per_second": 133.621,
716
+ "eval_steps_per_second": 4.182,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 2.872378243867757,
721
+ "grad_norm": 0.6100145578384399,
722
+ "learning_rate": 2.2776454444153328e-07,
723
+ "loss": 1.6978,
724
+ "step": 1010
725
+ },
726
+ {
727
+ "epoch": 2.900817632420903,
728
+ "grad_norm": 0.6016635894775391,
729
+ "learning_rate": 1.342297002141918e-07,
730
+ "loss": 1.6623,
731
+ "step": 1020
732
+ },
733
+ {
734
+ "epoch": 2.929257020974049,
735
+ "grad_norm": 0.5942381620407104,
736
+ "learning_rate": 6.523436919190773e-08,
737
+ "loss": 1.6999,
738
+ "step": 1030
739
+ },
740
+ {
741
+ "epoch": 2.957696409527195,
742
+ "grad_norm": 0.6014872789382935,
743
+ "learning_rate": 2.0846641436497726e-08,
744
+ "loss": 1.6816,
745
+ "step": 1040
746
+ },
747
+ {
748
+ "epoch": 2.9861357980803414,
749
+ "grad_norm": 0.6061838269233704,
750
+ "learning_rate": 1.1103222762542941e-09,
751
+ "loss": 1.7004,
752
+ "step": 1050
753
+ },
754
+ {
755
+ "epoch": 2.994667614646285,
756
+ "step": 1053,
757
+ "total_flos": 4.229776342129836e+18,
758
+ "train_loss": 1.8626822329427895,
759
+ "train_runtime": 5321.5336,
760
+ "train_samples_per_second": 50.737,
761
+ "train_steps_per_second": 0.198
762
+ }
763
+ ],
764
+ "logging_steps": 10,
765
+ "max_steps": 1053,
766
+ "num_input_tokens_seen": 0,
767
+ "num_train_epochs": 3,
768
+ "save_steps": 500,
769
+ "stateful_callbacks": {
770
+ "TrainerControl": {
771
+ "args": {
772
+ "should_epoch_stop": false,
773
+ "should_evaluate": false,
774
+ "should_log": false,
775
+ "should_save": true,
776
+ "should_training_stop": true
777
+ },
778
+ "attributes": {}
779
+ }
780
+ },
781
+ "total_flos": 4.229776342129836e+18,
782
+ "train_batch_size": 4,
783
+ "trial_name": null,
784
+ "trial_params": null
785
+ }
training_eval_loss.png ADDED
training_loss.png ADDED