sedrickkeh commited on
Commit
0cb4e72
·
verified ·
1 Parent(s): 1118340

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: mistralai/Mistral-7B-v0.1
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: hp_ablations_mistral_scheduler_linear_warmup0.05
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # hp_ablations_mistral_scheduler_linear_warmup0.05
17
 
18
- This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.0701
21
 
 
4
  base_model: mistralai/Mistral-7B-v0.1
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: hp_ablations_mistral_scheduler_linear_warmup0.05
 
16
 
17
  # hp_ablations_mistral_scheduler_linear_warmup0.05
18
 
19
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the mlfoundations-dev/oh-dcft-v3.1-gpt-4o-mini dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.0701
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9954430379746837,
3
+ "eval_loss": 0.0700838565826416,
4
+ "eval_runtime": 510.6467,
5
+ "eval_samples_per_second": 26.053,
6
+ "eval_steps_per_second": 0.407,
7
+ "total_flos": 2477170706350080.0,
8
+ "train_loss": 0.4888964453768779,
9
+ "train_runtime": 84815.3932,
10
+ "train_samples_per_second": 8.94,
11
+ "train_steps_per_second": 0.017
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9954430379746837,
3
+ "eval_loss": 0.0700838565826416,
4
+ "eval_runtime": 510.6467,
5
+ "eval_samples_per_second": 26.053,
6
+ "eval_steps_per_second": 0.407
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9954430379746837,
3
+ "total_flos": 2477170706350080.0,
4
+ "train_loss": 0.4888964453768779,
5
+ "train_runtime": 84815.3932,
6
+ "train_samples_per_second": 8.94,
7
+ "train_steps_per_second": 0.017
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1095 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9954430379746837,
5
+ "eval_steps": 500,
6
+ "global_step": 1479,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.020253164556962026,
13
+ "grad_norm": 4.742581199201631,
14
+ "learning_rate": 6.756756756756758e-07,
15
+ "loss": 0.8238,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.04050632911392405,
20
+ "grad_norm": 2.7089177182440225,
21
+ "learning_rate": 1.3513513513513515e-06,
22
+ "loss": 0.7256,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.060759493670886074,
27
+ "grad_norm": 1.6438330526897271,
28
+ "learning_rate": 2.0270270270270273e-06,
29
+ "loss": 0.6705,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.0810126582278481,
34
+ "grad_norm": 1.9649059006708485,
35
+ "learning_rate": 2.702702702702703e-06,
36
+ "loss": 0.6382,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.10126582278481013,
41
+ "grad_norm": 1.8319812554661556,
42
+ "learning_rate": 3.3783783783783788e-06,
43
+ "loss": 0.62,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.12151898734177215,
48
+ "grad_norm": 1.8563981143868757,
49
+ "learning_rate": 4.0540540540540545e-06,
50
+ "loss": 0.6097,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.14177215189873418,
55
+ "grad_norm": 2.0369557567075196,
56
+ "learning_rate": 4.72972972972973e-06,
57
+ "loss": 0.6039,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.1620253164556962,
62
+ "grad_norm": 3.933193428286484,
63
+ "learning_rate": 4.978647686832741e-06,
64
+ "loss": 0.5981,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.18227848101265823,
69
+ "grad_norm": 3.961786031844899,
70
+ "learning_rate": 4.943060498220641e-06,
71
+ "loss": 0.5957,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.20253164556962025,
76
+ "grad_norm": 3.016653386179453,
77
+ "learning_rate": 4.907473309608541e-06,
78
+ "loss": 0.5966,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.22278481012658227,
83
+ "grad_norm": 2.0911611009486664,
84
+ "learning_rate": 4.8718861209964415e-06,
85
+ "loss": 0.5874,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2430379746835443,
90
+ "grad_norm": 2.3577222635919877,
91
+ "learning_rate": 4.836298932384342e-06,
92
+ "loss": 0.5851,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.26329113924050634,
97
+ "grad_norm": 2.4189498060173826,
98
+ "learning_rate": 4.800711743772242e-06,
99
+ "loss": 0.5839,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.28354430379746837,
104
+ "grad_norm": 1.9667579738688012,
105
+ "learning_rate": 4.765124555160142e-06,
106
+ "loss": 0.573,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3037974683544304,
111
+ "grad_norm": 1.369959691403867,
112
+ "learning_rate": 4.7295373665480435e-06,
113
+ "loss": 0.5838,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.3240506329113924,
118
+ "grad_norm": 1.4080935903120366,
119
+ "learning_rate": 4.693950177935944e-06,
120
+ "loss": 0.5826,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.34430379746835443,
125
+ "grad_norm": 1.375777676837261,
126
+ "learning_rate": 4.658362989323843e-06,
127
+ "loss": 0.5827,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.36455696202531646,
132
+ "grad_norm": 1.6546345406257401,
133
+ "learning_rate": 4.622775800711744e-06,
134
+ "loss": 0.5783,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.3848101265822785,
139
+ "grad_norm": 2.138184576302769,
140
+ "learning_rate": 4.587188612099645e-06,
141
+ "loss": 0.578,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.4050632911392405,
146
+ "grad_norm": 1.5042114599648924,
147
+ "learning_rate": 4.551601423487545e-06,
148
+ "loss": 0.5713,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.4253164556962025,
153
+ "grad_norm": 2.34307337846717,
154
+ "learning_rate": 4.516014234875445e-06,
155
+ "loss": 0.5659,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.44556962025316454,
160
+ "grad_norm": 2.2965660467623983,
161
+ "learning_rate": 4.4804270462633455e-06,
162
+ "loss": 0.5711,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.46582278481012657,
167
+ "grad_norm": 2.2958689571859363,
168
+ "learning_rate": 4.444839857651246e-06,
169
+ "loss": 0.5693,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.4860759493670886,
174
+ "grad_norm": 2.417004138332344,
175
+ "learning_rate": 4.409252669039146e-06,
176
+ "loss": 0.5646,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5063291139240507,
181
+ "grad_norm": 2.12941087663038,
182
+ "learning_rate": 4.373665480427046e-06,
183
+ "loss": 0.5681,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.5265822784810127,
188
+ "grad_norm": 1.8555366173965417,
189
+ "learning_rate": 4.3380782918149475e-06,
190
+ "loss": 0.5543,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.5468354430379747,
195
+ "grad_norm": 1.5852520856985666,
196
+ "learning_rate": 4.302491103202847e-06,
197
+ "loss": 0.5639,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.5670886075949367,
202
+ "grad_norm": 1.460016106438118,
203
+ "learning_rate": 4.266903914590747e-06,
204
+ "loss": 0.5593,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.5873417721518988,
209
+ "grad_norm": 1.7121558558722028,
210
+ "learning_rate": 4.231316725978648e-06,
211
+ "loss": 0.5608,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.6075949367088608,
216
+ "grad_norm": 1.3067137647800322,
217
+ "learning_rate": 4.195729537366549e-06,
218
+ "loss": 0.5607,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.6278481012658228,
223
+ "grad_norm": 1.418773706013845,
224
+ "learning_rate": 4.160142348754449e-06,
225
+ "loss": 0.5666,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.6481012658227848,
230
+ "grad_norm": 1.4267499965709702,
231
+ "learning_rate": 4.124555160142349e-06,
232
+ "loss": 0.5629,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.6683544303797468,
237
+ "grad_norm": 1.3749578870885502,
238
+ "learning_rate": 4.0889679715302495e-06,
239
+ "loss": 0.5619,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.6886075949367089,
244
+ "grad_norm": 1.275893799768224,
245
+ "learning_rate": 4.05338078291815e-06,
246
+ "loss": 0.5611,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.7088607594936709,
251
+ "grad_norm": 1.2613374964537525,
252
+ "learning_rate": 4.01779359430605e-06,
253
+ "loss": 0.5625,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.7291139240506329,
258
+ "grad_norm": 1.282128305662025,
259
+ "learning_rate": 3.98220640569395e-06,
260
+ "loss": 0.5647,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.7493670886075949,
265
+ "grad_norm": 1.2183620482413273,
266
+ "learning_rate": 3.946619217081851e-06,
267
+ "loss": 0.557,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.769620253164557,
272
+ "grad_norm": 1.1816403948954313,
273
+ "learning_rate": 3.911032028469751e-06,
274
+ "loss": 0.5535,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.789873417721519,
279
+ "grad_norm": 1.343065511383417,
280
+ "learning_rate": 3.875444839857651e-06,
281
+ "loss": 0.5564,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.810126582278481,
286
+ "grad_norm": 1.1056411836705247,
287
+ "learning_rate": 3.839857651245552e-06,
288
+ "loss": 0.5562,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.830379746835443,
293
+ "grad_norm": 1.1617514638384954,
294
+ "learning_rate": 3.804270462633452e-06,
295
+ "loss": 0.5509,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.850632911392405,
300
+ "grad_norm": 1.2456514736887305,
301
+ "learning_rate": 3.7686832740213525e-06,
302
+ "loss": 0.5589,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.8708860759493671,
307
+ "grad_norm": 1.0959898492388527,
308
+ "learning_rate": 3.733096085409253e-06,
309
+ "loss": 0.5521,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.8911392405063291,
314
+ "grad_norm": 1.295665638247925,
315
+ "learning_rate": 3.6975088967971536e-06,
316
+ "loss": 0.5572,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.9113924050632911,
321
+ "grad_norm": 1.0535123677725715,
322
+ "learning_rate": 3.661921708185054e-06,
323
+ "loss": 0.5482,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.9316455696202531,
328
+ "grad_norm": 1.0683406426734499,
329
+ "learning_rate": 3.6263345195729537e-06,
330
+ "loss": 0.5525,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.9518987341772152,
335
+ "grad_norm": 1.253383608055468,
336
+ "learning_rate": 3.5907473309608544e-06,
337
+ "loss": 0.5533,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.9721518987341772,
342
+ "grad_norm": 1.2809747280599841,
343
+ "learning_rate": 3.5551601423487547e-06,
344
+ "loss": 0.5505,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.9924050632911392,
349
+ "grad_norm": 1.1325843807943288,
350
+ "learning_rate": 3.519572953736655e-06,
351
+ "loss": 0.5465,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.9984810126582279,
356
+ "eval_loss": 0.068679079413414,
357
+ "eval_runtime": 507.7278,
358
+ "eval_samples_per_second": 26.203,
359
+ "eval_steps_per_second": 0.41,
360
+ "step": 493
361
+ },
362
+ {
363
+ "epoch": 1.0126582278481013,
364
+ "grad_norm": 1.7325407692218586,
365
+ "learning_rate": 3.4839857651245557e-06,
366
+ "loss": 0.5125,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.0329113924050632,
371
+ "grad_norm": 1.9559729603669886,
372
+ "learning_rate": 3.4483985765124556e-06,
373
+ "loss": 0.4812,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.0531645569620254,
378
+ "grad_norm": 1.4932963289367234,
379
+ "learning_rate": 3.412811387900356e-06,
380
+ "loss": 0.4774,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.0734177215189873,
385
+ "grad_norm": 1.185384048798669,
386
+ "learning_rate": 3.3772241992882566e-06,
387
+ "loss": 0.4785,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.0936708860759494,
392
+ "grad_norm": 1.1972600486999827,
393
+ "learning_rate": 3.341637010676157e-06,
394
+ "loss": 0.4758,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.1139240506329113,
399
+ "grad_norm": 1.1460022054980692,
400
+ "learning_rate": 3.3060498220640576e-06,
401
+ "loss": 0.4778,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.1341772151898735,
406
+ "grad_norm": 1.1197932329811855,
407
+ "learning_rate": 3.2704626334519575e-06,
408
+ "loss": 0.4752,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.1544303797468354,
413
+ "grad_norm": 1.194566549573746,
414
+ "learning_rate": 3.2348754448398577e-06,
415
+ "loss": 0.4818,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.1746835443037975,
420
+ "grad_norm": 1.2784411634842332,
421
+ "learning_rate": 3.1992882562277585e-06,
422
+ "loss": 0.4798,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.1949367088607594,
427
+ "grad_norm": 1.123327911933,
428
+ "learning_rate": 3.1637010676156587e-06,
429
+ "loss": 0.4736,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.2151898734177216,
434
+ "grad_norm": 1.193089413979018,
435
+ "learning_rate": 3.128113879003559e-06,
436
+ "loss": 0.4804,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.2354430379746835,
441
+ "grad_norm": 1.1785101089717391,
442
+ "learning_rate": 3.0925266903914593e-06,
443
+ "loss": 0.4762,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.2556962025316456,
448
+ "grad_norm": 1.2080168650481182,
449
+ "learning_rate": 3.0569395017793596e-06,
450
+ "loss": 0.4798,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.2759493670886077,
455
+ "grad_norm": 1.3205102063888547,
456
+ "learning_rate": 3.02135231316726e-06,
457
+ "loss": 0.4796,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.2962025316455696,
462
+ "grad_norm": 1.3147319714200894,
463
+ "learning_rate": 2.9857651245551606e-06,
464
+ "loss": 0.4787,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.3164556962025316,
469
+ "grad_norm": 1.1490461889736139,
470
+ "learning_rate": 2.950177935943061e-06,
471
+ "loss": 0.482,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.3367088607594937,
476
+ "grad_norm": 1.3228852312739667,
477
+ "learning_rate": 2.9145907473309608e-06,
478
+ "loss": 0.4705,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.3569620253164558,
483
+ "grad_norm": 1.2638864667189769,
484
+ "learning_rate": 2.8790035587188615e-06,
485
+ "loss": 0.4839,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.3772151898734177,
490
+ "grad_norm": 1.1378222725461256,
491
+ "learning_rate": 2.8434163701067618e-06,
492
+ "loss": 0.4807,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.3974683544303796,
497
+ "grad_norm": 1.136931180029992,
498
+ "learning_rate": 2.8078291814946625e-06,
499
+ "loss": 0.4793,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.4177215189873418,
504
+ "grad_norm": 1.2382582028693814,
505
+ "learning_rate": 2.7722419928825624e-06,
506
+ "loss": 0.476,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.437974683544304,
511
+ "grad_norm": 1.2304522818128008,
512
+ "learning_rate": 2.7366548042704626e-06,
513
+ "loss": 0.4813,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.4582278481012658,
518
+ "grad_norm": 1.123509723565589,
519
+ "learning_rate": 2.7010676156583634e-06,
520
+ "loss": 0.4809,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.4784810126582277,
525
+ "grad_norm": 1.1387958282357735,
526
+ "learning_rate": 2.6654804270462636e-06,
527
+ "loss": 0.4833,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.4987341772151899,
532
+ "grad_norm": 1.118649730450664,
533
+ "learning_rate": 2.629893238434164e-06,
534
+ "loss": 0.4758,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.518987341772152,
539
+ "grad_norm": 1.222896592137077,
540
+ "learning_rate": 2.5943060498220642e-06,
541
+ "loss": 0.4779,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.539240506329114,
546
+ "grad_norm": 1.4074398237958103,
547
+ "learning_rate": 2.5587188612099645e-06,
548
+ "loss": 0.4788,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.5594936708860758,
553
+ "grad_norm": 1.1619300400485018,
554
+ "learning_rate": 2.523131672597865e-06,
555
+ "loss": 0.4772,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.579746835443038,
560
+ "grad_norm": 1.422184034548514,
561
+ "learning_rate": 2.4875444839857655e-06,
562
+ "loss": 0.4778,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.6,
567
+ "grad_norm": 1.2234336626816806,
568
+ "learning_rate": 2.451957295373666e-06,
569
+ "loss": 0.4792,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.620253164556962,
574
+ "grad_norm": 1.1332543250276066,
575
+ "learning_rate": 2.416370106761566e-06,
576
+ "loss": 0.4753,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.640506329113924,
581
+ "grad_norm": 1.2069589755433405,
582
+ "learning_rate": 2.3807829181494664e-06,
583
+ "loss": 0.4683,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.660759493670886,
588
+ "grad_norm": 1.106986539594195,
589
+ "learning_rate": 2.3451957295373667e-06,
590
+ "loss": 0.4773,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.6810126582278482,
595
+ "grad_norm": 1.1080989558058947,
596
+ "learning_rate": 2.3096085409252674e-06,
597
+ "loss": 0.472,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.70126582278481,
602
+ "grad_norm": 1.1905792829948112,
603
+ "learning_rate": 2.2740213523131673e-06,
604
+ "loss": 0.4734,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.721518987341772,
609
+ "grad_norm": 1.1263868733328295,
610
+ "learning_rate": 2.238434163701068e-06,
611
+ "loss": 0.4755,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.7417721518987341,
616
+ "grad_norm": 1.033205957803366,
617
+ "learning_rate": 2.2028469750889683e-06,
618
+ "loss": 0.475,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.7620253164556963,
623
+ "grad_norm": 1.127189794873212,
624
+ "learning_rate": 2.1672597864768685e-06,
625
+ "loss": 0.4785,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.7822784810126582,
630
+ "grad_norm": 1.1788992895844184,
631
+ "learning_rate": 2.131672597864769e-06,
632
+ "loss": 0.479,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.80253164556962,
637
+ "grad_norm": 1.186766190893949,
638
+ "learning_rate": 2.096085409252669e-06,
639
+ "loss": 0.4763,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 1.8227848101265822,
644
+ "grad_norm": 1.1866935971232218,
645
+ "learning_rate": 2.06049822064057e-06,
646
+ "loss": 0.4751,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 1.8430379746835444,
651
+ "grad_norm": 1.2254733687693131,
652
+ "learning_rate": 2.0249110320284697e-06,
653
+ "loss": 0.4767,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 1.8632911392405065,
658
+ "grad_norm": 1.0999643877142318,
659
+ "learning_rate": 1.9893238434163704e-06,
660
+ "loss": 0.475,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 1.8835443037974684,
665
+ "grad_norm": 1.1191194093430747,
666
+ "learning_rate": 1.9537366548042703e-06,
667
+ "loss": 0.4714,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 1.9037974683544303,
672
+ "grad_norm": 1.1342720462254603,
673
+ "learning_rate": 1.918149466192171e-06,
674
+ "loss": 0.4751,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 1.9240506329113924,
679
+ "grad_norm": 1.0241324911004035,
680
+ "learning_rate": 1.8825622775800715e-06,
681
+ "loss": 0.4759,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 1.9443037974683546,
686
+ "grad_norm": 1.1373301367884234,
687
+ "learning_rate": 1.8469750889679716e-06,
688
+ "loss": 0.4795,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 1.9645569620253165,
693
+ "grad_norm": 1.047662657429949,
694
+ "learning_rate": 1.811387900355872e-06,
695
+ "loss": 0.4778,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 1.9848101265822784,
700
+ "grad_norm": 1.0623876456662695,
701
+ "learning_rate": 1.7758007117437724e-06,
702
+ "loss": 0.4778,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 1.998987341772152,
707
+ "eval_loss": 0.06775479018688202,
708
+ "eval_runtime": 509.7296,
709
+ "eval_samples_per_second": 26.1,
710
+ "eval_steps_per_second": 0.408,
711
+ "step": 987
712
+ },
713
+ {
714
+ "epoch": 2.0050632911392405,
715
+ "grad_norm": 1.885410249625527,
716
+ "learning_rate": 1.7402135231316727e-06,
717
+ "loss": 0.456,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.0253164556962027,
722
+ "grad_norm": 1.4856083185892484,
723
+ "learning_rate": 1.7046263345195732e-06,
724
+ "loss": 0.4107,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.0455696202531644,
729
+ "grad_norm": 1.309133994163287,
730
+ "learning_rate": 1.6690391459074735e-06,
731
+ "loss": 0.4093,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.0658227848101265,
736
+ "grad_norm": 1.1938535418758685,
737
+ "learning_rate": 1.633451957295374e-06,
738
+ "loss": 0.4054,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.0860759493670886,
743
+ "grad_norm": 1.107962295104452,
744
+ "learning_rate": 1.597864768683274e-06,
745
+ "loss": 0.4053,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.1063291139240508,
750
+ "grad_norm": 1.2052202862963048,
751
+ "learning_rate": 1.5622775800711745e-06,
752
+ "loss": 0.4024,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.1265822784810124,
757
+ "grad_norm": 1.1688527474064736,
758
+ "learning_rate": 1.526690391459075e-06,
759
+ "loss": 0.411,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.1468354430379746,
764
+ "grad_norm": 1.159582820630888,
765
+ "learning_rate": 1.4911032028469751e-06,
766
+ "loss": 0.4056,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.1670886075949367,
771
+ "grad_norm": 1.1801911147741104,
772
+ "learning_rate": 1.4555160142348756e-06,
773
+ "loss": 0.4079,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.187341772151899,
778
+ "grad_norm": 1.1866560972006943,
779
+ "learning_rate": 1.419928825622776e-06,
780
+ "loss": 0.4061,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.207594936708861,
785
+ "grad_norm": 1.1271721136927093,
786
+ "learning_rate": 1.3843416370106764e-06,
787
+ "loss": 0.4035,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.2278481012658227,
792
+ "grad_norm": 1.148968106587725,
793
+ "learning_rate": 1.3487544483985765e-06,
794
+ "loss": 0.412,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.248101265822785,
799
+ "grad_norm": 1.17444595906901,
800
+ "learning_rate": 1.313167259786477e-06,
801
+ "loss": 0.4078,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.268354430379747,
806
+ "grad_norm": 1.241550247309238,
807
+ "learning_rate": 1.2775800711743775e-06,
808
+ "loss": 0.4091,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.2886075949367086,
813
+ "grad_norm": 1.1322921512574697,
814
+ "learning_rate": 1.2419928825622776e-06,
815
+ "loss": 0.41,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.3088607594936708,
820
+ "grad_norm": 1.1038688400048324,
821
+ "learning_rate": 1.206405693950178e-06,
822
+ "loss": 0.4052,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.329113924050633,
827
+ "grad_norm": 1.1471785710291027,
828
+ "learning_rate": 1.1708185053380784e-06,
829
+ "loss": 0.4116,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.349367088607595,
834
+ "grad_norm": 1.1748513395840992,
835
+ "learning_rate": 1.1352313167259789e-06,
836
+ "loss": 0.4104,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.369620253164557,
841
+ "grad_norm": 1.1687832541778012,
842
+ "learning_rate": 1.0996441281138791e-06,
843
+ "loss": 0.4103,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.389873417721519,
848
+ "grad_norm": 1.2153707569716299,
849
+ "learning_rate": 1.0640569395017794e-06,
850
+ "loss": 0.4102,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.410126582278481,
855
+ "grad_norm": 1.2043915036892876,
856
+ "learning_rate": 1.0284697508896797e-06,
857
+ "loss": 0.4093,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.430379746835443,
862
+ "grad_norm": 1.1154060559605754,
863
+ "learning_rate": 9.928825622775802e-07,
864
+ "loss": 0.4077,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.4506329113924052,
869
+ "grad_norm": 1.1249500649551158,
870
+ "learning_rate": 9.572953736654805e-07,
871
+ "loss": 0.4067,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.470886075949367,
876
+ "grad_norm": 1.1373080270061644,
877
+ "learning_rate": 9.217081850533809e-07,
878
+ "loss": 0.4094,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.491139240506329,
883
+ "grad_norm": 1.2009940834893456,
884
+ "learning_rate": 8.861209964412812e-07,
885
+ "loss": 0.4081,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.511392405063291,
890
+ "grad_norm": 1.1805699402158027,
891
+ "learning_rate": 8.505338078291815e-07,
892
+ "loss": 0.412,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.5316455696202533,
897
+ "grad_norm": 1.1650230538613433,
898
+ "learning_rate": 8.14946619217082e-07,
899
+ "loss": 0.4053,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.5518987341772155,
904
+ "grad_norm": 1.1148652618881494,
905
+ "learning_rate": 7.793594306049823e-07,
906
+ "loss": 0.4003,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.572151898734177,
911
+ "grad_norm": 1.1203790714411743,
912
+ "learning_rate": 7.437722419928827e-07,
913
+ "loss": 0.4071,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.5924050632911393,
918
+ "grad_norm": 1.1540738131106723,
919
+ "learning_rate": 7.08185053380783e-07,
920
+ "loss": 0.4088,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 2.6126582278481014,
925
+ "grad_norm": 1.1564560350860578,
926
+ "learning_rate": 6.725978647686833e-07,
927
+ "loss": 0.4136,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.632911392405063,
932
+ "grad_norm": 1.1502008333731566,
933
+ "learning_rate": 6.370106761565837e-07,
934
+ "loss": 0.4074,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.6531645569620252,
939
+ "grad_norm": 1.084342294241873,
940
+ "learning_rate": 6.01423487544484e-07,
941
+ "loss": 0.4068,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.6734177215189874,
946
+ "grad_norm": 1.0975552769986037,
947
+ "learning_rate": 5.658362989323843e-07,
948
+ "loss": 0.408,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 2.6936708860759495,
953
+ "grad_norm": 1.1097418341903917,
954
+ "learning_rate": 5.302491103202847e-07,
955
+ "loss": 0.4103,
956
+ "step": 1330
957
+ },
958
+ {
959
+ "epoch": 2.7139240506329116,
960
+ "grad_norm": 1.158009570288043,
961
+ "learning_rate": 4.946619217081851e-07,
962
+ "loss": 0.4086,
963
+ "step": 1340
964
+ },
965
+ {
966
+ "epoch": 2.7341772151898733,
967
+ "grad_norm": 1.082032638132166,
968
+ "learning_rate": 4.5907473309608547e-07,
969
+ "loss": 0.4067,
970
+ "step": 1350
971
+ },
972
+ {
973
+ "epoch": 2.7544303797468355,
974
+ "grad_norm": 1.0910706188001162,
975
+ "learning_rate": 4.234875444839858e-07,
976
+ "loss": 0.4025,
977
+ "step": 1360
978
+ },
979
+ {
980
+ "epoch": 2.7746835443037976,
981
+ "grad_norm": 1.1340308917418294,
982
+ "learning_rate": 3.8790035587188615e-07,
983
+ "loss": 0.4074,
984
+ "step": 1370
985
+ },
986
+ {
987
+ "epoch": 2.7949367088607593,
988
+ "grad_norm": 1.0972217494408671,
989
+ "learning_rate": 3.5231316725978655e-07,
990
+ "loss": 0.4059,
991
+ "step": 1380
992
+ },
993
+ {
994
+ "epoch": 2.8151898734177214,
995
+ "grad_norm": 1.1023421234405986,
996
+ "learning_rate": 3.1672597864768684e-07,
997
+ "loss": 0.408,
998
+ "step": 1390
999
+ },
1000
+ {
1001
+ "epoch": 2.8354430379746836,
1002
+ "grad_norm": 1.076348844263821,
1003
+ "learning_rate": 2.811387900355872e-07,
1004
+ "loss": 0.4008,
1005
+ "step": 1400
1006
+ },
1007
+ {
1008
+ "epoch": 2.8556962025316457,
1009
+ "grad_norm": 1.0727773827003273,
1010
+ "learning_rate": 2.455516014234876e-07,
1011
+ "loss": 0.4095,
1012
+ "step": 1410
1013
+ },
1014
+ {
1015
+ "epoch": 2.875949367088608,
1016
+ "grad_norm": 1.094488859203195,
1017
+ "learning_rate": 2.0996441281138792e-07,
1018
+ "loss": 0.4044,
1019
+ "step": 1420
1020
+ },
1021
+ {
1022
+ "epoch": 2.8962025316455695,
1023
+ "grad_norm": 1.1022638043714281,
1024
+ "learning_rate": 1.743772241992883e-07,
1025
+ "loss": 0.4053,
1026
+ "step": 1430
1027
+ },
1028
+ {
1029
+ "epoch": 2.9164556962025316,
1030
+ "grad_norm": 1.056288437679614,
1031
+ "learning_rate": 1.3879003558718863e-07,
1032
+ "loss": 0.407,
1033
+ "step": 1440
1034
+ },
1035
+ {
1036
+ "epoch": 2.9367088607594938,
1037
+ "grad_norm": 1.1311358899472859,
1038
+ "learning_rate": 1.0320284697508897e-07,
1039
+ "loss": 0.4152,
1040
+ "step": 1450
1041
+ },
1042
+ {
1043
+ "epoch": 2.9569620253164555,
1044
+ "grad_norm": 1.0577864076415648,
1045
+ "learning_rate": 6.761565836298933e-08,
1046
+ "loss": 0.4032,
1047
+ "step": 1460
1048
+ },
1049
+ {
1050
+ "epoch": 2.9772151898734176,
1051
+ "grad_norm": 1.0913097767086777,
1052
+ "learning_rate": 3.202846975088968e-08,
1053
+ "loss": 0.4067,
1054
+ "step": 1470
1055
+ },
1056
+ {
1057
+ "epoch": 2.9954430379746837,
1058
+ "eval_loss": 0.0700838565826416,
1059
+ "eval_runtime": 511.013,
1060
+ "eval_samples_per_second": 26.035,
1061
+ "eval_steps_per_second": 0.407,
1062
+ "step": 1479
1063
+ },
1064
+ {
1065
+ "epoch": 2.9954430379746837,
1066
+ "step": 1479,
1067
+ "total_flos": 2477170706350080.0,
1068
+ "train_loss": 0.4888964453768779,
1069
+ "train_runtime": 84815.3932,
1070
+ "train_samples_per_second": 8.94,
1071
+ "train_steps_per_second": 0.017
1072
+ }
1073
+ ],
1074
+ "logging_steps": 10,
1075
+ "max_steps": 1479,
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": 2477170706350080.0,
1092
+ "train_batch_size": 8,
1093
+ "trial_name": null,
1094
+ "trial_params": null
1095
+ }
training_eval_loss.png ADDED
training_loss.png ADDED