sedrickkeh commited on
Commit
771e22c
·
verified ·
1 Parent(s): c6d51b5

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: top_13_ranking_stackexchange
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # top_13_ranking_stackexchange
17
 
18
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.7935
21
 
 
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: top_13_ranking_stackexchange
 
16
 
17
  # top_13_ranking_stackexchange
18
 
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/top_13_ranking_stackexchange dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.7935
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996871741397289,
3
+ "eval_loss": 0.7934871912002563,
4
+ "eval_runtime": 511.9706,
5
+ "eval_samples_per_second": 25.238,
6
+ "eval_steps_per_second": 0.395,
7
+ "total_flos": 2406819142041600.0,
8
+ "train_loss": 0.7739501513784431,
9
+ "train_runtime": 84617.9298,
10
+ "train_samples_per_second": 8.703,
11
+ "train_steps_per_second": 0.017
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996871741397289,
3
+ "eval_loss": 0.7934871912002563,
4
+ "eval_runtime": 511.9706,
5
+ "eval_samples_per_second": 25.238,
6
+ "eval_steps_per_second": 0.395
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996871741397289,
3
+ "total_flos": 2406819142041600.0,
4
+ "train_loss": 0.7739501513784431,
5
+ "train_runtime": 84617.9298,
6
+ "train_samples_per_second": 8.703,
7
+ "train_steps_per_second": 0.017
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1067 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.996871741397289,
5
+ "eval_steps": 500,
6
+ "global_step": 1437,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.020855057351407715,
13
+ "grad_norm": 23.292541836109518,
14
+ "learning_rate": 5e-06,
15
+ "loss": 1.058,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.04171011470281543,
20
+ "grad_norm": 1.8444543970999359,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.9571,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.06256517205422316,
27
+ "grad_norm": 2.3601682539820206,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.9306,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.08342022940563086,
34
+ "grad_norm": 1.5228442928811052,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.9114,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.10427528675703858,
41
+ "grad_norm": 1.0380683585757473,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.8914,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.1251303441084463,
48
+ "grad_norm": 2.607870506748757,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.8813,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.145985401459854,
55
+ "grad_norm": 1.135440860340382,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.873,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.16684045881126172,
62
+ "grad_norm": 1.5314230383194811,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.8648,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.18769551616266944,
69
+ "grad_norm": 1.191321725126771,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.8557,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.20855057351407716,
76
+ "grad_norm": 0.8966237449653851,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.8552,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.22940563086548488,
83
+ "grad_norm": 0.8329312402745882,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.8452,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2502606882168926,
90
+ "grad_norm": 0.6887837863764799,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.8473,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.2711157455683003,
97
+ "grad_norm": 0.5586690368831195,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.8469,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.291970802919708,
104
+ "grad_norm": 0.5603617843797895,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.8419,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.31282586027111575,
111
+ "grad_norm": 0.6421219835657508,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.838,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.33368091762252344,
118
+ "grad_norm": 0.6230076979493112,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.8386,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.3545359749739312,
125
+ "grad_norm": 0.7839779910445314,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.8388,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.3753910323253389,
132
+ "grad_norm": 0.7493442852757208,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.8397,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.3962460896767466,
139
+ "grad_norm": 0.5814041052759352,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.8323,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.4171011470281543,
146
+ "grad_norm": 0.6300228137380017,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.8298,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.43795620437956206,
153
+ "grad_norm": 0.5636230176981548,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.8356,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.45881126173096975,
160
+ "grad_norm": 0.7908527390096476,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.8279,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.4796663190823775,
167
+ "grad_norm": 0.595200937080435,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.8263,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5005213764337852,
174
+ "grad_norm": 0.7675029006003027,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.8246,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5213764337851929,
181
+ "grad_norm": 0.6352165366458988,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.8231,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.5422314911366006,
188
+ "grad_norm": 0.7073190898153328,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.8208,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.5630865484880083,
195
+ "grad_norm": 0.626287882243934,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.821,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.583941605839416,
202
+ "grad_norm": 0.6897085554274643,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.8197,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.6047966631908238,
209
+ "grad_norm": 0.5407250019832245,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.8188,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.6256517205422315,
216
+ "grad_norm": 0.7961611092737484,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.8203,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.6465067778936392,
223
+ "grad_norm": 0.6416129777930016,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.8169,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.6673618352450469,
230
+ "grad_norm": 0.8877371521398986,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.8195,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.6882168925964547,
237
+ "grad_norm": 0.5637078899529996,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.8155,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.7090719499478624,
244
+ "grad_norm": 0.6162994025511069,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.8134,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.7299270072992701,
251
+ "grad_norm": 0.6274925211049429,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.8191,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.7507820646506778,
258
+ "grad_norm": 0.47554028354411854,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.8137,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.7716371220020855,
265
+ "grad_norm": 0.7364107778965528,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.8137,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.7924921793534933,
272
+ "grad_norm": 0.5387935911559895,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.8132,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.8133472367049009,
279
+ "grad_norm": 0.653190218549883,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.8135,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.8342022940563086,
286
+ "grad_norm": 0.5854373414156873,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.812,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.8550573514077163,
293
+ "grad_norm": 0.5267893296142316,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.8142,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.8759124087591241,
300
+ "grad_norm": 0.6645124564599936,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.8055,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.8967674661105318,
307
+ "grad_norm": 0.7119869644519043,
308
+ "learning_rate": 5e-06,
309
+ "loss": 0.8107,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.9176225234619395,
314
+ "grad_norm": 0.5835795125930624,
315
+ "learning_rate": 5e-06,
316
+ "loss": 0.8046,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.9384775808133472,
321
+ "grad_norm": 0.7229576221148301,
322
+ "learning_rate": 5e-06,
323
+ "loss": 0.8114,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.959332638164755,
328
+ "grad_norm": 0.4979281463667173,
329
+ "learning_rate": 5e-06,
330
+ "loss": 0.8085,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.9801876955161627,
335
+ "grad_norm": 0.7630249389366001,
336
+ "learning_rate": 5e-06,
337
+ "loss": 0.8049,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.9989572471324296,
342
+ "eval_loss": 0.8077179789543152,
343
+ "eval_runtime": 508.9353,
344
+ "eval_samples_per_second": 25.388,
345
+ "eval_steps_per_second": 0.397,
346
+ "step": 479
347
+ },
348
+ {
349
+ "epoch": 1.0010427528675705,
350
+ "grad_norm": 0.9794940422687983,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.8452,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.0218978102189782,
357
+ "grad_norm": 0.7307911661230689,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.7627,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.0427528675703859,
364
+ "grad_norm": 0.6035063157192385,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.7631,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.0636079249217936,
371
+ "grad_norm": 0.5103542458227681,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.7592,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.0844629822732013,
378
+ "grad_norm": 0.6509093187706326,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.7646,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.105318039624609,
385
+ "grad_norm": 0.5243282773682663,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.7623,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.1261730969760166,
392
+ "grad_norm": 0.5059428734291583,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.7605,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.1470281543274243,
399
+ "grad_norm": 0.5440857998105081,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.7632,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.167883211678832,
406
+ "grad_norm": 0.6125935114881672,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.7603,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.1887382690302397,
413
+ "grad_norm": 0.663818253738316,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.7608,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.2095933263816476,
420
+ "grad_norm": 0.5078369936190525,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.7638,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.2304483837330553,
427
+ "grad_norm": 0.508989568319443,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.7674,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.251303441084463,
434
+ "grad_norm": 0.483682589877922,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.7649,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.2721584984358707,
441
+ "grad_norm": 0.4761829121492384,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.7659,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.2930135557872784,
448
+ "grad_norm": 0.5647667401130814,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.7668,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.313868613138686,
455
+ "grad_norm": 0.6452289804237794,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.7591,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.3347236704900938,
462
+ "grad_norm": 0.5975511480256017,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.7637,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.3555787278415017,
469
+ "grad_norm": 0.5786109716986558,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.7621,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.3764337851929094,
476
+ "grad_norm": 0.5162003159982257,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.7616,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.397288842544317,
483
+ "grad_norm": 0.8301389411240245,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.7679,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.4181438998957248,
490
+ "grad_norm": 0.6643589136117731,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.7626,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.4389989572471324,
497
+ "grad_norm": 0.5575453250895497,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.7636,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.4598540145985401,
504
+ "grad_norm": 0.6311203856335321,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.7616,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.4807090719499478,
511
+ "grad_norm": 0.5162506907296919,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.7647,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.5015641293013555,
518
+ "grad_norm": 0.5886130199524026,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.7598,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.5224191866527632,
525
+ "grad_norm": 0.6696049559875061,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.7629,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.543274244004171,
532
+ "grad_norm": 0.5799722177117436,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.7656,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.5641293013555786,
539
+ "grad_norm": 0.5370703772665276,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.7634,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.5849843587069863,
546
+ "grad_norm": 0.5298044539792373,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.7624,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.6058394160583942,
553
+ "grad_norm": 0.5184639411381536,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.7601,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.6266944734098019,
560
+ "grad_norm": 0.48600399732107674,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.7598,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.6475495307612096,
567
+ "grad_norm": 0.5686370144176348,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.7582,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.6684045881126173,
574
+ "grad_norm": 0.5132646507076349,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.7621,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.6892596454640252,
581
+ "grad_norm": 0.4731379428525773,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.7624,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.7101147028154329,
588
+ "grad_norm": 0.5785964102816669,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.7621,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.7309697601668406,
595
+ "grad_norm": 0.522780904551143,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.7615,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.7518248175182483,
602
+ "grad_norm": 0.4966543168759053,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.759,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.772679874869656,
609
+ "grad_norm": 0.5481286697337492,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.7623,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.7935349322210636,
616
+ "grad_norm": 0.567267871896373,
617
+ "learning_rate": 5e-06,
618
+ "loss": 0.7561,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.8143899895724713,
623
+ "grad_norm": 0.49959155217976836,
624
+ "learning_rate": 5e-06,
625
+ "loss": 0.763,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.835245046923879,
630
+ "grad_norm": 0.577604273627036,
631
+ "learning_rate": 5e-06,
632
+ "loss": 0.7581,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.8561001042752867,
637
+ "grad_norm": 0.5751847723051718,
638
+ "learning_rate": 5e-06,
639
+ "loss": 0.7639,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 1.8769551616266944,
644
+ "grad_norm": 0.5107772226236036,
645
+ "learning_rate": 5e-06,
646
+ "loss": 0.7588,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 1.897810218978102,
651
+ "grad_norm": 0.5819488319800581,
652
+ "learning_rate": 5e-06,
653
+ "loss": 0.7676,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 1.9186652763295098,
658
+ "grad_norm": 0.4842327798393644,
659
+ "learning_rate": 5e-06,
660
+ "loss": 0.7621,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 1.9395203336809175,
665
+ "grad_norm": 0.4932679101447122,
666
+ "learning_rate": 5e-06,
667
+ "loss": 0.7582,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 1.9603753910323254,
672
+ "grad_norm": 0.4470310666364999,
673
+ "learning_rate": 5e-06,
674
+ "loss": 0.7607,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 1.981230448383733,
679
+ "grad_norm": 0.4770177687209944,
680
+ "learning_rate": 5e-06,
681
+ "loss": 0.7573,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 2.0,
686
+ "eval_loss": 0.7946059703826904,
687
+ "eval_runtime": 508.0791,
688
+ "eval_samples_per_second": 25.431,
689
+ "eval_steps_per_second": 0.398,
690
+ "step": 959
691
+ },
692
+ {
693
+ "epoch": 2.002085505735141,
694
+ "grad_norm": 1.032414598918905,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.7947,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.0229405630865487,
701
+ "grad_norm": 0.6854847007975855,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.7156,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.0437956204379564,
708
+ "grad_norm": 0.5340454379181093,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.7127,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.064650677789364,
715
+ "grad_norm": 0.6846712179907463,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.7126,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.0855057351407718,
722
+ "grad_norm": 0.6601525652275346,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.714,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.1063607924921794,
729
+ "grad_norm": 0.5087564518975344,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.7131,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.127215849843587,
736
+ "grad_norm": 0.5717589267694552,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.7144,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.148070907194995,
743
+ "grad_norm": 0.6240703812823883,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.7145,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.1689259645464025,
750
+ "grad_norm": 0.5700642714485155,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.7112,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.18978102189781,
757
+ "grad_norm": 0.5407339761292322,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.7182,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.210636079249218,
764
+ "grad_norm": 0.612301975168881,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.7189,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.2314911366006256,
771
+ "grad_norm": 0.5060820644021846,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.7222,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.2523461939520333,
778
+ "grad_norm": 0.6530049157665407,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.7188,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.273201251303441,
785
+ "grad_norm": 0.6861670806344601,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.7196,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.2940563086548487,
792
+ "grad_norm": 0.6641025315790597,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.7162,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.3149113660062564,
799
+ "grad_norm": 0.5829858270987744,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.7157,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.335766423357664,
806
+ "grad_norm": 0.5365315836646892,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.7164,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.356621480709072,
813
+ "grad_norm": 0.5742395072896596,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.7204,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.3774765380604794,
820
+ "grad_norm": 0.6686102534842351,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.7172,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.3983315954118876,
827
+ "grad_norm": 0.5623753120923718,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.7192,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.4191866527632953,
834
+ "grad_norm": 0.5595969776381658,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.7182,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.440041710114703,
841
+ "grad_norm": 0.5265170403517665,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.7147,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.4608967674661106,
848
+ "grad_norm": 0.6149297351724455,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.7156,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.4817518248175183,
855
+ "grad_norm": 0.5126528944575456,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.7203,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.502606882168926,
862
+ "grad_norm": 0.6126480417488975,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.721,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.5234619395203337,
869
+ "grad_norm": 0.6298522226894485,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.7186,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.5443169968717414,
876
+ "grad_norm": 0.7016975692629115,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.7181,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.565172054223149,
883
+ "grad_norm": 0.5235129334510048,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.7185,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.586027111574557,
890
+ "grad_norm": 0.6772917599095623,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.7176,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.6068821689259645,
897
+ "grad_norm": 0.605127895215746,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.7162,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.627737226277372,
904
+ "grad_norm": 0.6817953131541243,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.7203,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.64859228362878,
911
+ "grad_norm": 0.570350382612417,
912
+ "learning_rate": 5e-06,
913
+ "loss": 0.7167,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.6694473409801875,
918
+ "grad_norm": 0.5585743640080483,
919
+ "learning_rate": 5e-06,
920
+ "loss": 0.7188,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 2.6903023983315952,
925
+ "grad_norm": 0.5480849125671278,
926
+ "learning_rate": 5e-06,
927
+ "loss": 0.7166,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.7111574556830034,
932
+ "grad_norm": 0.4990381771165155,
933
+ "learning_rate": 5e-06,
934
+ "loss": 0.7189,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.7320125130344106,
939
+ "grad_norm": 0.5035279549109828,
940
+ "learning_rate": 5e-06,
941
+ "loss": 0.7195,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.7528675703858188,
946
+ "grad_norm": 0.5308531250097626,
947
+ "learning_rate": 5e-06,
948
+ "loss": 0.7188,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 2.7737226277372264,
953
+ "grad_norm": 0.5685033810774691,
954
+ "learning_rate": 5e-06,
955
+ "loss": 0.7216,
956
+ "step": 1330
957
+ },
958
+ {
959
+ "epoch": 2.794577685088634,
960
+ "grad_norm": 0.6063980901472933,
961
+ "learning_rate": 5e-06,
962
+ "loss": 0.7208,
963
+ "step": 1340
964
+ },
965
+ {
966
+ "epoch": 2.815432742440042,
967
+ "grad_norm": 0.5655728920483287,
968
+ "learning_rate": 5e-06,
969
+ "loss": 0.7155,
970
+ "step": 1350
971
+ },
972
+ {
973
+ "epoch": 2.8362877997914495,
974
+ "grad_norm": 0.5857668680906439,
975
+ "learning_rate": 5e-06,
976
+ "loss": 0.7233,
977
+ "step": 1360
978
+ },
979
+ {
980
+ "epoch": 2.857142857142857,
981
+ "grad_norm": 0.5372669669303354,
982
+ "learning_rate": 5e-06,
983
+ "loss": 0.7219,
984
+ "step": 1370
985
+ },
986
+ {
987
+ "epoch": 2.877997914494265,
988
+ "grad_norm": 0.5464301975129316,
989
+ "learning_rate": 5e-06,
990
+ "loss": 0.718,
991
+ "step": 1380
992
+ },
993
+ {
994
+ "epoch": 2.8988529718456726,
995
+ "grad_norm": 0.5087739900733502,
996
+ "learning_rate": 5e-06,
997
+ "loss": 0.7196,
998
+ "step": 1390
999
+ },
1000
+ {
1001
+ "epoch": 2.9197080291970803,
1002
+ "grad_norm": 0.5561518718385676,
1003
+ "learning_rate": 5e-06,
1004
+ "loss": 0.7195,
1005
+ "step": 1400
1006
+ },
1007
+ {
1008
+ "epoch": 2.940563086548488,
1009
+ "grad_norm": 0.6788170550296148,
1010
+ "learning_rate": 5e-06,
1011
+ "loss": 0.715,
1012
+ "step": 1410
1013
+ },
1014
+ {
1015
+ "epoch": 2.9614181438998957,
1016
+ "grad_norm": 0.5770145877498132,
1017
+ "learning_rate": 5e-06,
1018
+ "loss": 0.7213,
1019
+ "step": 1420
1020
+ },
1021
+ {
1022
+ "epoch": 2.9822732012513034,
1023
+ "grad_norm": 0.5587054163866665,
1024
+ "learning_rate": 5e-06,
1025
+ "loss": 0.7202,
1026
+ "step": 1430
1027
+ },
1028
+ {
1029
+ "epoch": 2.996871741397289,
1030
+ "eval_loss": 0.7934871912002563,
1031
+ "eval_runtime": 509.6159,
1032
+ "eval_samples_per_second": 25.354,
1033
+ "eval_steps_per_second": 0.396,
1034
+ "step": 1437
1035
+ },
1036
+ {
1037
+ "epoch": 2.996871741397289,
1038
+ "step": 1437,
1039
+ "total_flos": 2406819142041600.0,
1040
+ "train_loss": 0.7739501513784431,
1041
+ "train_runtime": 84617.9298,
1042
+ "train_samples_per_second": 8.703,
1043
+ "train_steps_per_second": 0.017
1044
+ }
1045
+ ],
1046
+ "logging_steps": 10,
1047
+ "max_steps": 1437,
1048
+ "num_input_tokens_seen": 0,
1049
+ "num_train_epochs": 3,
1050
+ "save_steps": 500,
1051
+ "stateful_callbacks": {
1052
+ "TrainerControl": {
1053
+ "args": {
1054
+ "should_epoch_stop": false,
1055
+ "should_evaluate": false,
1056
+ "should_log": false,
1057
+ "should_save": true,
1058
+ "should_training_stop": true
1059
+ },
1060
+ "attributes": {}
1061
+ }
1062
+ },
1063
+ "total_flos": 2406819142041600.0,
1064
+ "train_batch_size": 8,
1065
+ "trial_name": null,
1066
+ "trial_params": null
1067
+ }
training_eval_loss.png ADDED
training_loss.png ADDED