sedrickkeh commited on
Commit
d7a5915
·
verified ·
1 Parent(s): 4235f1b

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: mistralai/Mistral-7B-v0.3
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-405b-v2dummytesting
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-405b-v2dummytesting
17
 
18
- This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.0340
21
 
 
4
  base_model: mistralai/Mistral-7B-v0.3
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-405b-v2dummytesting
 
16
 
17
  # mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-405b-v2dummytesting
18
 
19
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the mlfoundations-dev/oh-dcft-v3.1-llama-3.1-405b dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.0340
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996178343949045,
3
+ "eval_loss": 0.03400026634335518,
4
+ "eval_runtime": 320.4428,
5
+ "eval_samples_per_second": 41.259,
6
+ "eval_steps_per_second": 0.646,
7
+ "total_flos": 2462095371141120.0,
8
+ "train_loss": 0.19492420712295844,
9
+ "train_runtime": 47387.4935,
10
+ "train_samples_per_second": 15.902,
11
+ "train_steps_per_second": 0.031
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996178343949045,
3
+ "eval_loss": 0.03400026634335518,
4
+ "eval_runtime": 320.4428,
5
+ "eval_samples_per_second": 41.259,
6
+ "eval_steps_per_second": 0.646
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996178343949045,
3
+ "total_flos": 2462095371141120.0,
4
+ "train_loss": 0.19492420712295844,
5
+ "train_runtime": 47387.4935,
6
+ "train_samples_per_second": 15.902,
7
+ "train_steps_per_second": 0.031
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1095 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.996178343949045,
5
+ "eval_steps": 500,
6
+ "global_step": 1470,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02038216560509554,
13
+ "grad_norm": 7.3640559222825805,
14
+ "learning_rate": 6.756756756756758e-07,
15
+ "loss": 0.5218,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.04076433121019108,
20
+ "grad_norm": 1.9597929093946138,
21
+ "learning_rate": 1.3513513513513515e-06,
22
+ "loss": 0.4442,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.061146496815286625,
27
+ "grad_norm": 1.366556978476813,
28
+ "learning_rate": 2.0270270270270273e-06,
29
+ "loss": 0.3968,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.08152866242038216,
34
+ "grad_norm": 1.5343177845993703,
35
+ "learning_rate": 2.702702702702703e-06,
36
+ "loss": 0.3717,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.10191082802547771,
41
+ "grad_norm": 1.5235479557468576,
42
+ "learning_rate": 3.3783783783783788e-06,
43
+ "loss": 0.3573,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.12229299363057325,
48
+ "grad_norm": 1.5210742465140967,
49
+ "learning_rate": 4.0540540540540545e-06,
50
+ "loss": 0.3447,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.14267515923566879,
55
+ "grad_norm": 1.4560884271057324,
56
+ "learning_rate": 4.72972972972973e-06,
57
+ "loss": 0.336,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.16305732484076432,
62
+ "grad_norm": 1.3653484275359835,
63
+ "learning_rate": 4.999794894487084e-06,
64
+ "loss": 0.3285,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.18343949044585986,
69
+ "grad_norm": 1.5231339353031599,
70
+ "learning_rate": 4.998541607324267e-06,
71
+ "loss": 0.3226,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.20382165605095542,
76
+ "grad_norm": 1.34927340034923,
77
+ "learning_rate": 4.996149614425655e-06,
78
+ "loss": 0.3234,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.22420382165605096,
83
+ "grad_norm": 1.3383365926789343,
84
+ "learning_rate": 4.992620127143546e-06,
85
+ "loss": 0.3159,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2445859872611465,
90
+ "grad_norm": 1.279507514764783,
91
+ "learning_rate": 4.987954932879781e-06,
92
+ "loss": 0.3155,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.26496815286624203,
97
+ "grad_norm": 1.2523662594214922,
98
+ "learning_rate": 4.982156394180578e-06,
99
+ "loss": 0.3089,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.28535031847133757,
104
+ "grad_norm": 1.5508610615211977,
105
+ "learning_rate": 4.975227447540084e-06,
106
+ "loss": 0.3081,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3057324840764331,
111
+ "grad_norm": 1.5577467324922831,
112
+ "learning_rate": 4.967171601913286e-06,
113
+ "loss": 0.2999,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.32611464968152865,
118
+ "grad_norm": 1.4024580607747494,
119
+ "learning_rate": 4.957992936939001e-06,
120
+ "loss": 0.3014,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.3464968152866242,
125
+ "grad_norm": 1.397084384608207,
126
+ "learning_rate": 4.947696100873868e-06,
127
+ "loss": 0.3003,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.3668789808917197,
132
+ "grad_norm": 1.2608713430895078,
133
+ "learning_rate": 4.936286308238376e-06,
134
+ "loss": 0.2962,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.3872611464968153,
139
+ "grad_norm": 1.4602055977294606,
140
+ "learning_rate": 4.923769337176137e-06,
141
+ "loss": 0.2949,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.40764331210191085,
146
+ "grad_norm": 1.3574725912947458,
147
+ "learning_rate": 4.910151526527711e-06,
148
+ "loss": 0.2913,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.4280254777070064,
153
+ "grad_norm": 1.3255996037154008,
154
+ "learning_rate": 4.895439772620496e-06,
155
+ "loss": 0.2909,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.4484076433121019,
160
+ "grad_norm": 1.209522300925548,
161
+ "learning_rate": 4.879641525776294e-06,
162
+ "loss": 0.2874,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.46878980891719746,
167
+ "grad_norm": 1.3036670455198982,
168
+ "learning_rate": 4.8627647865383185e-06,
169
+ "loss": 0.2887,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.489171974522293,
174
+ "grad_norm": 1.4420527770171871,
175
+ "learning_rate": 4.844818101619563e-06,
176
+ "loss": 0.2893,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5095541401273885,
181
+ "grad_norm": 1.5008539741569011,
182
+ "learning_rate": 4.825810559574577e-06,
183
+ "loss": 0.2831,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.5299363057324841,
188
+ "grad_norm": 1.1363685690254681,
189
+ "learning_rate": 4.805751786196844e-06,
190
+ "loss": 0.2781,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.5503184713375796,
195
+ "grad_norm": 1.3886569509043332,
196
+ "learning_rate": 4.784651939644088e-06,
197
+ "loss": 0.2812,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.5707006369426751,
202
+ "grad_norm": 1.208611174951388,
203
+ "learning_rate": 4.762521705293985e-06,
204
+ "loss": 0.2839,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.5910828025477707,
209
+ "grad_norm": 1.163584249987649,
210
+ "learning_rate": 4.739372290332876e-06,
211
+ "loss": 0.2772,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.6114649681528662,
216
+ "grad_norm": 1.2694569325884293,
217
+ "learning_rate": 4.715215418080228e-06,
218
+ "loss": 0.2774,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.6318471337579618,
223
+ "grad_norm": 1.1438695314175629,
224
+ "learning_rate": 4.690063322051714e-06,
225
+ "loss": 0.279,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.6522292993630573,
230
+ "grad_norm": 1.1455759975487239,
231
+ "learning_rate": 4.6639287397639175e-06,
232
+ "loss": 0.2762,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.6726114649681528,
237
+ "grad_norm": 1.44840804645582,
238
+ "learning_rate": 4.636824906283807e-06,
239
+ "loss": 0.2752,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.6929936305732484,
244
+ "grad_norm": 1.1318324006100882,
245
+ "learning_rate": 4.608765547526235e-06,
246
+ "loss": 0.2696,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.7133757961783439,
251
+ "grad_norm": 1.4105199522926484,
252
+ "learning_rate": 4.5797648733028665e-06,
253
+ "loss": 0.2714,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.7337579617834394,
258
+ "grad_norm": 1.1828019600051816,
259
+ "learning_rate": 4.54983757012605e-06,
260
+ "loss": 0.2675,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.7541401273885351,
265
+ "grad_norm": 1.1209568717673557,
266
+ "learning_rate": 4.518998793771276e-06,
267
+ "loss": 0.2698,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.7745222929936306,
272
+ "grad_norm": 1.3068840388262846,
273
+ "learning_rate": 4.487264161601997e-06,
274
+ "loss": 0.2666,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.7949044585987262,
279
+ "grad_norm": 1.40475775183392,
280
+ "learning_rate": 4.454649744660687e-06,
281
+ "loss": 0.2687,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.8152866242038217,
286
+ "grad_norm": 1.064147060517822,
287
+ "learning_rate": 4.4211720595301474e-06,
288
+ "loss": 0.2668,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.8356687898089172,
293
+ "grad_norm": 1.273665530555761,
294
+ "learning_rate": 4.386848059969186e-06,
295
+ "loss": 0.2655,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.8560509554140128,
300
+ "grad_norm": 1.2644936064921932,
301
+ "learning_rate": 4.351695128326899e-06,
302
+ "loss": 0.267,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.8764331210191083,
307
+ "grad_norm": 1.1524074004874778,
308
+ "learning_rate": 4.315731066739907e-06,
309
+ "loss": 0.2621,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.8968152866242038,
314
+ "grad_norm": 1.2734388359258864,
315
+ "learning_rate": 4.278974088117002e-06,
316
+ "loss": 0.2646,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.9171974522292994,
321
+ "grad_norm": 1.1535683141057886,
322
+ "learning_rate": 4.241442806915776e-06,
323
+ "loss": 0.2622,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.9375796178343949,
328
+ "grad_norm": 1.1270411566076795,
329
+ "learning_rate": 4.203156229715885e-06,
330
+ "loss": 0.2647,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.9579617834394905,
335
+ "grad_norm": 1.1492915594780593,
336
+ "learning_rate": 4.164133745593752e-06,
337
+ "loss": 0.2604,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.978343949044586,
342
+ "grad_norm": 1.1077787065388747,
343
+ "learning_rate": 4.1243951163035515e-06,
344
+ "loss": 0.2589,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.9987261146496815,
349
+ "grad_norm": 1.04425073531816,
350
+ "learning_rate": 4.083960466269468e-06,
351
+ "loss": 0.2567,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.9987261146496815,
356
+ "eval_loss": 0.032234665006399155,
357
+ "eval_runtime": 329.9438,
358
+ "eval_samples_per_second": 40.07,
359
+ "eval_steps_per_second": 0.627,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.019108280254777,
364
+ "grad_norm": 1.4452926570009665,
365
+ "learning_rate": 4.042850272394296e-06,
366
+ "loss": 0.1917,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.0394904458598726,
371
+ "grad_norm": 1.1692376024393614,
372
+ "learning_rate": 4.001085353689527e-06,
373
+ "loss": 0.1828,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.0598726114649681,
378
+ "grad_norm": 1.2077457145616015,
379
+ "learning_rate": 3.958686860732198e-06,
380
+ "loss": 0.1823,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.0802547770700637,
385
+ "grad_norm": 1.1829681497737912,
386
+ "learning_rate": 3.915676264953819e-06,
387
+ "loss": 0.1824,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.1006369426751592,
392
+ "grad_norm": 1.212198809480288,
393
+ "learning_rate": 3.8720753477668174e-06,
394
+ "loss": 0.1835,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.1210191082802548,
399
+ "grad_norm": 1.4514789562338672,
400
+ "learning_rate": 3.8279061895340065e-06,
401
+ "loss": 0.1838,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.1414012738853503,
406
+ "grad_norm": 1.2292597073935565,
407
+ "learning_rate": 3.783191158386649e-06,
408
+ "loss": 0.1799,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.1617834394904458,
413
+ "grad_norm": 1.091816482106913,
414
+ "learning_rate": 3.7379528988968004e-06,
415
+ "loss": 0.1797,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.1821656050955414,
420
+ "grad_norm": 1.1348703066662176,
421
+ "learning_rate": 3.6922143206096463e-06,
422
+ "loss": 0.1814,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.202547770700637,
427
+ "grad_norm": 1.2696934307053864,
428
+ "learning_rate": 3.645998586441664e-06,
429
+ "loss": 0.1814,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.2229299363057324,
434
+ "grad_norm": 1.1409724326559059,
435
+ "learning_rate": 3.5993291009504583e-06,
436
+ "loss": 0.1828,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.243312101910828,
441
+ "grad_norm": 1.2402396526237325,
442
+ "learning_rate": 3.5522294984822414e-06,
443
+ "loss": 0.1842,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.2636942675159235,
448
+ "grad_norm": 1.289864855038617,
449
+ "learning_rate": 3.504723631202927e-06,
450
+ "loss": 0.1824,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.284076433121019,
455
+ "grad_norm": 1.1569779602515864,
456
+ "learning_rate": 3.4568355570189293e-06,
457
+ "loss": 0.1833,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.3044585987261146,
462
+ "grad_norm": 1.1664095681219364,
463
+ "learning_rate": 3.4085895273937616e-06,
464
+ "loss": 0.1822,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.3248407643312101,
469
+ "grad_norm": 1.1062062982676109,
470
+ "learning_rate": 3.3600099750666186e-06,
471
+ "loss": 0.1801,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.3452229299363057,
476
+ "grad_norm": 1.0870703803021455,
477
+ "learning_rate": 3.3111215016791615e-06,
478
+ "loss": 0.1811,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.3656050955414012,
483
+ "grad_norm": 1.070032061400707,
484
+ "learning_rate": 3.2619488653167524e-06,
485
+ "loss": 0.1816,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.385987261146497,
490
+ "grad_norm": 1.3367455282881715,
491
+ "learning_rate": 3.2125169679704816e-06,
492
+ "loss": 0.1813,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.4063694267515925,
497
+ "grad_norm": 1.187247916290702,
498
+ "learning_rate": 3.1628508429263055e-06,
499
+ "loss": 0.1779,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.426751592356688,
504
+ "grad_norm": 1.1091917590396907,
505
+ "learning_rate": 3.112975642087696e-06,
506
+ "loss": 0.1802,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.4471337579617836,
511
+ "grad_norm": 1.123698745710789,
512
+ "learning_rate": 3.0629166232382244e-06,
513
+ "loss": 0.1814,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.467515923566879,
518
+ "grad_norm": 1.1042223426325448,
519
+ "learning_rate": 3.012699137250523e-06,
520
+ "loss": 0.1824,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.4878980891719746,
525
+ "grad_norm": 1.050554441234522,
526
+ "learning_rate": 2.962348615248099e-06,
527
+ "loss": 0.1796,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.5082802547770702,
532
+ "grad_norm": 1.2199041359267095,
533
+ "learning_rate": 2.9118905557265194e-06,
534
+ "loss": 0.181,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.5286624203821657,
539
+ "grad_norm": 1.1808278399972332,
540
+ "learning_rate": 2.861350511640466e-06,
541
+ "loss": 0.1803,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.5490445859872612,
546
+ "grad_norm": 1.0852792148234631,
547
+ "learning_rate": 2.81075407746322e-06,
548
+ "loss": 0.1798,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.5694267515923568,
553
+ "grad_norm": 1.1397500179113436,
554
+ "learning_rate": 2.760126876225113e-06,
555
+ "loss": 0.1794,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.5898089171974523,
560
+ "grad_norm": 1.0868979078693752,
561
+ "learning_rate": 2.7094945465375237e-06,
562
+ "loss": 0.1784,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.6101910828025479,
567
+ "grad_norm": 1.185135837873844,
568
+ "learning_rate": 2.6588827296089765e-06,
569
+ "loss": 0.1796,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.6305732484076434,
574
+ "grad_norm": 1.0667742842947785,
575
+ "learning_rate": 2.608317056259933e-06,
576
+ "loss": 0.1778,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.650955414012739,
581
+ "grad_norm": 1.1263879048027836,
582
+ "learning_rate": 2.557823133942836e-06,
583
+ "loss": 0.1818,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.6713375796178345,
588
+ "grad_norm": 1.164600608354229,
589
+ "learning_rate": 2.507426533773994e-06,
590
+ "loss": 0.1818,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.69171974522293,
595
+ "grad_norm": 1.1434673428111148,
596
+ "learning_rate": 2.4571527775838637e-06,
597
+ "loss": 0.1785,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.7121019108280255,
602
+ "grad_norm": 1.1080982356818843,
603
+ "learning_rate": 2.407027324992293e-06,
604
+ "loss": 0.1767,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.732484076433121,
609
+ "grad_norm": 1.1236879328083964,
610
+ "learning_rate": 2.357075560515267e-06,
611
+ "loss": 0.1764,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.7528662420382166,
616
+ "grad_norm": 1.07775485257901,
617
+ "learning_rate": 2.3073227807096903e-06,
618
+ "loss": 0.1766,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.7732484076433122,
623
+ "grad_norm": 1.0978325229339359,
624
+ "learning_rate": 2.2577941813627137e-06,
625
+ "loss": 0.1776,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.7936305732484077,
630
+ "grad_norm": 1.0266269472342355,
631
+ "learning_rate": 2.2085148447320886e-06,
632
+ "loss": 0.1763,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.8140127388535032,
637
+ "grad_norm": 1.075781931859834,
638
+ "learning_rate": 2.1595097268440215e-06,
639
+ "loss": 0.1761,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 1.8343949044585988,
644
+ "grad_norm": 1.1486032295285624,
645
+ "learning_rate": 2.110803644854949e-06,
646
+ "loss": 0.1756,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 1.8547770700636943,
651
+ "grad_norm": 1.0561746481203738,
652
+ "learning_rate": 2.062421264483641e-06,
653
+ "loss": 0.1754,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 1.8751592356687898,
658
+ "grad_norm": 1.1673017813466813,
659
+ "learning_rate": 2.0143870875199952e-06,
660
+ "loss": 0.1764,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 1.8955414012738854,
665
+ "grad_norm": 1.1439217602579965,
666
+ "learning_rate": 1.9667254394168497e-06,
667
+ "loss": 0.1742,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 1.915923566878981,
672
+ "grad_norm": 1.1584206154770065,
673
+ "learning_rate": 1.9194604569710933e-06,
674
+ "loss": 0.1795,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 1.9363057324840764,
679
+ "grad_norm": 1.0480077245390713,
680
+ "learning_rate": 1.8726160761003172e-06,
681
+ "loss": 0.1748,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 1.956687898089172,
686
+ "grad_norm": 1.1162273428280736,
687
+ "learning_rate": 1.8262160197211954e-06,
688
+ "loss": 0.1761,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 1.9770700636942675,
693
+ "grad_norm": 1.1160462930422237,
694
+ "learning_rate": 1.780283785735729e-06,
695
+ "loss": 0.174,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 1.997452229299363,
700
+ "grad_norm": 1.062423344058906,
701
+ "learning_rate": 1.734842635131455e-06,
702
+ "loss": 0.1749,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 1.9994904458598726,
707
+ "eval_loss": 0.030678020790219307,
708
+ "eval_runtime": 338.9213,
709
+ "eval_samples_per_second": 39.009,
710
+ "eval_steps_per_second": 0.611,
711
+ "step": 981
712
+ },
713
+ {
714
+ "epoch": 2.0178343949044586,
715
+ "grad_norm": 1.4150638555592978,
716
+ "learning_rate": 1.689915580201614e-06,
717
+ "loss": 0.1179,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.038216560509554,
722
+ "grad_norm": 1.230931823658668,
723
+ "learning_rate": 1.6455253728912765e-06,
724
+ "loss": 0.106,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.0585987261146497,
729
+ "grad_norm": 1.078514779623916,
730
+ "learning_rate": 1.6016944932753123e-06,
731
+ "loss": 0.1055,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.078980891719745,
736
+ "grad_norm": 1.1594743845163793,
737
+ "learning_rate": 1.558445138174033e-06,
738
+ "loss": 0.1068,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.0993630573248407,
743
+ "grad_norm": 1.1071569450433687,
744
+ "learning_rate": 1.5157992099122896e-06,
745
+ "loss": 0.1061,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.1197452229299363,
750
+ "grad_norm": 1.1427254217014393,
751
+ "learning_rate": 1.4737783052277017e-06,
752
+ "loss": 0.1054,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.140127388535032,
757
+ "grad_norm": 1.1862335656291514,
758
+ "learning_rate": 1.432403704333643e-06,
759
+ "loss": 0.1056,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.1605095541401274,
764
+ "grad_norm": 1.1251247878141808,
765
+ "learning_rate": 1.3916963601425293e-06,
766
+ "loss": 0.1058,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.180891719745223,
771
+ "grad_norm": 1.1084298272819255,
772
+ "learning_rate": 1.3516768876548404e-06,
773
+ "loss": 0.1068,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.2012738853503184,
778
+ "grad_norm": 1.1823405832071576,
779
+ "learning_rate": 1.3123655535192873e-06,
780
+ "loss": 0.1059,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.221656050955414,
785
+ "grad_norm": 1.1309238431972,
786
+ "learning_rate": 1.2737822657693763e-06,
787
+ "loss": 0.106,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.2420382165605095,
792
+ "grad_norm": 1.1830449008249062,
793
+ "learning_rate": 1.235946563741594e-06,
794
+ "loss": 0.106,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.262420382165605,
799
+ "grad_norm": 1.1465025990749025,
800
+ "learning_rate": 1.1988776081802975e-06,
801
+ "loss": 0.1049,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.2828025477707006,
806
+ "grad_norm": 1.183957689341336,
807
+ "learning_rate": 1.1625941715343404e-06,
808
+ "loss": 0.1054,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.303184713375796,
813
+ "grad_norm": 1.1333744410647408,
814
+ "learning_rate": 1.1271146284503326e-06,
815
+ "loss": 0.1055,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.3235668789808916,
820
+ "grad_norm": 1.1879433920175209,
821
+ "learning_rate": 1.0924569464673593e-06,
822
+ "loss": 0.1055,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.343949044585987,
827
+ "grad_norm": 1.146900632019417,
828
+ "learning_rate": 1.0586386769178644e-06,
829
+ "loss": 0.106,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.3643312101910827,
834
+ "grad_norm": 1.132036422681508,
835
+ "learning_rate": 1.0256769460393087e-06,
836
+ "loss": 0.1056,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.3847133757961783,
841
+ "grad_norm": 1.0853627031360338,
842
+ "learning_rate": 9.935884463011108e-07,
843
+ "loss": 0.1045,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.405095541401274,
848
+ "grad_norm": 1.1322288498135122,
849
+ "learning_rate": 9.623894279512455e-07,
850
+ "loss": 0.1054,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.4254777070063693,
855
+ "grad_norm": 1.0894698960326725,
856
+ "learning_rate": 9.320956907868051e-07,
857
+ "loss": 0.1056,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.445859872611465,
862
+ "grad_norm": 1.1474536034439595,
863
+ "learning_rate": 9.027225761526653e-07,
864
+ "loss": 0.1043,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.4662420382165604,
869
+ "grad_norm": 1.1143668115183167,
870
+ "learning_rate": 8.742849591723315e-07,
871
+ "loss": 0.1065,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.486624203821656,
876
+ "grad_norm": 1.1060933833835738,
877
+ "learning_rate": 8.467972412148767e-07,
878
+ "loss": 0.1049,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.5070063694267515,
883
+ "grad_norm": 1.1258549382443122,
884
+ "learning_rate": 8.202733426018117e-07,
885
+ "loss": 0.1039,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.527388535031847,
890
+ "grad_norm": 1.1747748534183018,
891
+ "learning_rate": 7.947266955575562e-07,
892
+ "loss": 0.1061,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.5477707006369426,
897
+ "grad_norm": 1.1215150862371803,
898
+ "learning_rate": 7.701702374070962e-07,
899
+ "loss": 0.1054,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.568152866242038,
904
+ "grad_norm": 1.1567778163976754,
905
+ "learning_rate": 7.466164040242694e-07,
906
+ "loss": 0.104,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.5885350318471336,
911
+ "grad_norm": 1.1338966482025068,
912
+ "learning_rate": 7.24077123533992e-07,
913
+ "loss": 0.1048,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.608917197452229,
918
+ "grad_norm": 1.0699364001483016,
919
+ "learning_rate": 7.025638102716238e-07,
920
+ "loss": 0.1046,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 2.6292993630573247,
925
+ "grad_norm": 1.085055290267827,
926
+ "learning_rate": 6.820873590025216e-07,
927
+ "loss": 0.1042,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.6496815286624202,
932
+ "grad_norm": 1.085099277831117,
933
+ "learning_rate": 6.626581394047174e-07,
934
+ "loss": 0.1043,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.6700636942675158,
939
+ "grad_norm": 1.1741130632881172,
940
+ "learning_rate": 6.442859908175084e-07,
941
+ "loss": 0.1053,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.6904458598726113,
946
+ "grad_norm": 1.086080275618794,
947
+ "learning_rate": 6.269802172586215e-07,
948
+ "loss": 0.1055,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 2.710828025477707,
953
+ "grad_norm": 1.1054751783303671,
954
+ "learning_rate": 6.107495827124764e-07,
955
+ "loss": 0.1041,
956
+ "step": 1330
957
+ },
958
+ {
959
+ "epoch": 2.7312101910828024,
960
+ "grad_norm": 1.0983120768210253,
961
+ "learning_rate": 5.956023066919313e-07,
962
+ "loss": 0.1026,
963
+ "step": 1340
964
+ },
965
+ {
966
+ "epoch": 2.7515923566878984,
967
+ "grad_norm": 1.109557569000835,
968
+ "learning_rate": 5.815460600757599e-07,
969
+ "loss": 0.1037,
970
+ "step": 1350
971
+ },
972
+ {
973
+ "epoch": 2.771974522292994,
974
+ "grad_norm": 1.1282912667207265,
975
+ "learning_rate": 5.685879612239649e-07,
976
+ "loss": 0.1057,
977
+ "step": 1360
978
+ },
979
+ {
980
+ "epoch": 2.7923566878980894,
981
+ "grad_norm": 1.0863425429571905,
982
+ "learning_rate": 5.567345723729062e-07,
983
+ "loss": 0.1047,
984
+ "step": 1370
985
+ },
986
+ {
987
+ "epoch": 2.812738853503185,
988
+ "grad_norm": 1.0546515835835897,
989
+ "learning_rate": 5.459918963120477e-07,
990
+ "loss": 0.1042,
991
+ "step": 1380
992
+ },
993
+ {
994
+ "epoch": 2.8331210191082805,
995
+ "grad_norm": 1.0748377352085035,
996
+ "learning_rate": 5.363653733440328e-07,
997
+ "loss": 0.1057,
998
+ "step": 1390
999
+ },
1000
+ {
1001
+ "epoch": 2.853503184713376,
1002
+ "grad_norm": 1.0869969236098675,
1003
+ "learning_rate": 5.278598785296044e-07,
1004
+ "loss": 0.105,
1005
+ "step": 1400
1006
+ },
1007
+ {
1008
+ "epoch": 2.8738853503184716,
1009
+ "grad_norm": 1.0525214029938958,
1010
+ "learning_rate": 5.2047971921878e-07,
1011
+ "loss": 0.1039,
1012
+ "step": 1410
1013
+ },
1014
+ {
1015
+ "epoch": 2.894267515923567,
1016
+ "grad_norm": 1.0788702761102988,
1017
+ "learning_rate": 5.142286328695235e-07,
1018
+ "loss": 0.105,
1019
+ "step": 1420
1020
+ },
1021
+ {
1022
+ "epoch": 2.9146496815286627,
1023
+ "grad_norm": 1.0984819174990816,
1024
+ "learning_rate": 5.091097851550239e-07,
1025
+ "loss": 0.1032,
1026
+ "step": 1430
1027
+ },
1028
+ {
1029
+ "epoch": 2.935031847133758,
1030
+ "grad_norm": 1.072624641241873,
1031
+ "learning_rate": 5.051257683605363e-07,
1032
+ "loss": 0.1036,
1033
+ "step": 1440
1034
+ },
1035
+ {
1036
+ "epoch": 2.9554140127388537,
1037
+ "grad_norm": 1.116104824847335,
1038
+ "learning_rate": 5.022786000705963e-07,
1039
+ "loss": 0.1027,
1040
+ "step": 1450
1041
+ },
1042
+ {
1043
+ "epoch": 2.9757961783439493,
1044
+ "grad_norm": 1.0814921699434623,
1045
+ "learning_rate": 5.005697221472769e-07,
1046
+ "loss": 0.1029,
1047
+ "step": 1460
1048
+ },
1049
+ {
1050
+ "epoch": 2.996178343949045,
1051
+ "grad_norm": 1.0668548799222997,
1052
+ "learning_rate": 5e-07,
1053
+ "loss": 0.1027,
1054
+ "step": 1470
1055
+ },
1056
+ {
1057
+ "epoch": 2.996178343949045,
1058
+ "eval_loss": 0.03400026634335518,
1059
+ "eval_runtime": 335.9006,
1060
+ "eval_samples_per_second": 39.36,
1061
+ "eval_steps_per_second": 0.616,
1062
+ "step": 1470
1063
+ },
1064
+ {
1065
+ "epoch": 2.996178343949045,
1066
+ "step": 1470,
1067
+ "total_flos": 2462095371141120.0,
1068
+ "train_loss": 0.19492420712295844,
1069
+ "train_runtime": 47387.4935,
1070
+ "train_samples_per_second": 15.902,
1071
+ "train_steps_per_second": 0.031
1072
+ }
1073
+ ],
1074
+ "logging_steps": 10,
1075
+ "max_steps": 1470,
1076
+ "num_input_tokens_seen": 0,
1077
+ "num_train_epochs": 3,
1078
+ "save_steps": 500,
1079
+ "stateful_callbacks": {
1080
+ "TrainerControl": {
1081
+ "args": {
1082
+ "should_epoch_stop": false,
1083
+ "should_evaluate": false,
1084
+ "should_log": false,
1085
+ "should_save": true,
1086
+ "should_training_stop": true
1087
+ },
1088
+ "attributes": {}
1089
+ }
1090
+ },
1091
+ "total_flos": 2462095371141120.0,
1092
+ "train_batch_size": 8,
1093
+ "trial_name": null,
1094
+ "trial_params": null
1095
+ }
training_eval_loss.png ADDED
training_loss.png ADDED