yakazimir commited on
Commit
bd58940
·
verified ·
1 Parent(s): acda86f

Model save

Browse files
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: HuggingFaceTB/SmolLM2-1.7B
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: smollm_1B_ultrachat
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # smollm_1B_ultrachat
20
+
21
+ This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-1.7B](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 1.2975
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 2e-05
43
+ - train_batch_size: 8
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 8
48
+ - gradient_accumulation_steps: 4
49
+ - total_train_batch_size: 256
50
+ - total_eval_batch_size: 64
51
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
52
+ - lr_scheduler_type: cosine
53
+ - lr_scheduler_warmup_ratio: 0.1
54
+ - num_epochs: 1
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss |
59
+ |:-------------:|:------:|:----:|:---------------:|
60
+ | 1.2996 | 0.4073 | 200 | 1.3281 |
61
+ | 1.2961 | 0.8147 | 400 | 1.2975 |
62
+
63
+
64
+ ### Framework versions
65
+
66
+ - Transformers 4.48.3
67
+ - Pytorch 2.2.2+cu121
68
+ - Datasets 2.18.0
69
+ - Tokenizers 0.21.0
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 154920007237632.0,
4
+ "train_loss": 1.4646558688760047,
5
+ "train_runtime": 2127.1747,
6
+ "train_samples": 207864,
7
+ "train_samples_per_second": 59.07,
8
+ "train_steps_per_second": 0.231
9
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 2,
6
+ "transformers_version": "4.48.3"
7
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 154920007237632.0,
4
+ "train_loss": 1.4646558688760047,
5
+ "train_runtime": 2127.1747,
6
+ "train_samples": 207864,
7
+ "train_samples_per_second": 59.07,
8
+ "train_steps_per_second": 0.231
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,751 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 200,
6
+ "global_step": 491,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.002036659877800407,
13
+ "grad_norm": 42.73765709178892,
14
+ "learning_rate": 4.0000000000000003e-07,
15
+ "loss": 4.2137,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.010183299389002037,
20
+ "grad_norm": 43.823126465212304,
21
+ "learning_rate": 2.0000000000000003e-06,
22
+ "loss": 4.2378,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.020366598778004074,
27
+ "grad_norm": 42.78292182490666,
28
+ "learning_rate": 4.000000000000001e-06,
29
+ "loss": 4.1532,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.03054989816700611,
34
+ "grad_norm": 26.956865206535298,
35
+ "learning_rate": 6e-06,
36
+ "loss": 3.7244,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.04073319755600815,
41
+ "grad_norm": 12.021358068486181,
42
+ "learning_rate": 8.000000000000001e-06,
43
+ "loss": 3.0576,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.05091649694501019,
48
+ "grad_norm": 10.364947442111331,
49
+ "learning_rate": 1e-05,
50
+ "loss": 2.5815,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.06109979633401222,
55
+ "grad_norm": 4.530399924565578,
56
+ "learning_rate": 1.2e-05,
57
+ "loss": 2.2026,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.07128309572301425,
62
+ "grad_norm": 3.2260177605870783,
63
+ "learning_rate": 1.4e-05,
64
+ "loss": 1.9904,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.0814663951120163,
69
+ "grad_norm": 2.12005760250828,
70
+ "learning_rate": 1.6000000000000003e-05,
71
+ "loss": 1.8236,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.09164969450101833,
76
+ "grad_norm": 1.8468068964995024,
77
+ "learning_rate": 1.8e-05,
78
+ "loss": 1.6897,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.10183299389002037,
83
+ "grad_norm": 1.5429747637898503,
84
+ "learning_rate": 2e-05,
85
+ "loss": 1.6026,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.1120162932790224,
90
+ "grad_norm": 1.1896025193659396,
91
+ "learning_rate": 1.9993657117550972e-05,
92
+ "loss": 1.5431,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.12219959266802444,
97
+ "grad_norm": 1.0406079144844387,
98
+ "learning_rate": 1.9974636516635436e-05,
99
+ "loss": 1.4948,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.13238289205702647,
104
+ "grad_norm": 0.9826731697259301,
105
+ "learning_rate": 1.994296232634054e-05,
106
+ "loss": 1.4551,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.1425661914460285,
111
+ "grad_norm": 0.8350404902447894,
112
+ "learning_rate": 1.9898674727799418e-05,
113
+ "loss": 1.4548,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.15274949083503056,
118
+ "grad_norm": 0.8022790771564833,
119
+ "learning_rate": 1.9841829903218377e-05,
120
+ "loss": 1.45,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.1629327902240326,
125
+ "grad_norm": 0.8137274794499317,
126
+ "learning_rate": 1.977249996460544e-05,
127
+ "loss": 1.4206,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.17311608961303462,
132
+ "grad_norm": 0.6702892273400702,
133
+ "learning_rate": 1.969077286229078e-05,
134
+ "loss": 1.4068,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.18329938900203666,
139
+ "grad_norm": 0.6712397854919697,
140
+ "learning_rate": 1.959675227335497e-05,
141
+ "loss": 1.4027,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.1934826883910387,
146
+ "grad_norm": 0.6017420494268232,
147
+ "learning_rate": 1.949055747010669e-05,
148
+ "loss": 1.4036,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.20366598778004075,
153
+ "grad_norm": 0.5946907262324206,
154
+ "learning_rate": 1.937232316877668e-05,
155
+ "loss": 1.4021,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.21384928716904278,
160
+ "grad_norm": 0.584049807273413,
161
+ "learning_rate": 1.9242199358619897e-05,
162
+ "loss": 1.3773,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.2240325865580448,
167
+ "grad_norm": 0.5468171068137498,
168
+ "learning_rate": 1.9100351111642666e-05,
169
+ "loss": 1.3824,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.23421588594704684,
174
+ "grad_norm": 0.5631810496676463,
175
+ "learning_rate": 1.894695837319623e-05,
176
+ "loss": 1.3716,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.24439918533604887,
181
+ "grad_norm": 0.5579171191759417,
182
+ "learning_rate": 1.8782215733702286e-05,
183
+ "loss": 1.3722,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.2545824847250509,
188
+ "grad_norm": 0.5126218869264142,
189
+ "learning_rate": 1.8606332181800165e-05,
190
+ "loss": 1.3523,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.26476578411405294,
195
+ "grad_norm": 0.5007973223598261,
196
+ "learning_rate": 1.841953083922875e-05,
197
+ "loss": 1.3451,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.27494908350305497,
202
+ "grad_norm": 0.5048172409656928,
203
+ "learning_rate": 1.8222048677779495e-05,
204
+ "loss": 1.3534,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.285132382892057,
209
+ "grad_norm": 0.4725169108639936,
210
+ "learning_rate": 1.8014136218679566e-05,
211
+ "loss": 1.3473,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.2953156822810591,
216
+ "grad_norm": 0.48611682020845765,
217
+ "learning_rate": 1.779605721478652e-05,
218
+ "loss": 1.3373,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.3054989816700611,
223
+ "grad_norm": 0.45300491076813515,
224
+ "learning_rate": 1.756808831599762e-05,
225
+ "loss": 1.3513,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.31568228105906315,
230
+ "grad_norm": 0.4382778646436873,
231
+ "learning_rate": 1.7330518718298263e-05,
232
+ "loss": 1.3342,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.3258655804480652,
237
+ "grad_norm": 0.4427941208449195,
238
+ "learning_rate": 1.7083649796894798e-05,
239
+ "loss": 1.3162,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.3360488798370672,
244
+ "grad_norm": 0.416296697128989,
245
+ "learning_rate": 1.6827794723896968e-05,
246
+ "loss": 1.3468,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.34623217922606925,
251
+ "grad_norm": 0.43327582549066207,
252
+ "learning_rate": 1.6563278071035182e-05,
253
+ "loss": 1.3413,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.3564154786150713,
258
+ "grad_norm": 0.39987863733708096,
259
+ "learning_rate": 1.6290435397916426e-05,
260
+ "loss": 1.3275,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.3665987780040733,
265
+ "grad_norm": 0.4107034128984309,
266
+ "learning_rate": 1.6009612826341226e-05,
267
+ "loss": 1.3165,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.37678207739307534,
272
+ "grad_norm": 0.39958981244827124,
273
+ "learning_rate": 1.5721166601221697e-05,
274
+ "loss": 1.3238,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.3869653767820774,
279
+ "grad_norm": 0.40620805351711664,
280
+ "learning_rate": 1.5425462638657597e-05,
281
+ "loss": 1.3265,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.3971486761710794,
286
+ "grad_norm": 0.3930588952515658,
287
+ "learning_rate": 1.5122876061743772e-05,
288
+ "loss": 1.3346,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.4073319755600815,
293
+ "grad_norm": 0.3455411843725847,
294
+ "learning_rate": 1.4813790724697832e-05,
295
+ "loss": 1.2996,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.4073319755600815,
300
+ "eval_loss": 1.3280781507492065,
301
+ "eval_runtime": 58.3849,
302
+ "eval_samples_per_second": 238.195,
303
+ "eval_steps_per_second": 3.734,
304
+ "step": 200
305
+ },
306
+ {
307
+ "epoch": 0.4175152749490835,
308
+ "grad_norm": 0.38779728978307215,
309
+ "learning_rate": 1.4498598725911693e-05,
310
+ "loss": 1.3155,
311
+ "step": 205
312
+ },
313
+ {
314
+ "epoch": 0.42769857433808556,
315
+ "grad_norm": 0.34758190751014123,
316
+ "learning_rate": 1.4177699910544793e-05,
317
+ "loss": 1.3226,
318
+ "step": 210
319
+ },
320
+ {
321
+ "epoch": 0.4378818737270876,
322
+ "grad_norm": 0.347052132116147,
323
+ "learning_rate": 1.3851501363289907e-05,
324
+ "loss": 1.3042,
325
+ "step": 215
326
+ },
327
+ {
328
+ "epoch": 0.4480651731160896,
329
+ "grad_norm": 0.35329848914934636,
330
+ "learning_rate": 1.3520416891955101e-05,
331
+ "loss": 1.3198,
332
+ "step": 220
333
+ },
334
+ {
335
+ "epoch": 0.45824847250509165,
336
+ "grad_norm": 0.3689489097870586,
337
+ "learning_rate": 1.3184866502516846e-05,
338
+ "loss": 1.3295,
339
+ "step": 225
340
+ },
341
+ {
342
+ "epoch": 0.4684317718940937,
343
+ "grad_norm": 0.3591427089290735,
344
+ "learning_rate": 1.2845275866310325e-05,
345
+ "loss": 1.3026,
346
+ "step": 230
347
+ },
348
+ {
349
+ "epoch": 0.4786150712830957,
350
+ "grad_norm": 0.33192490822778736,
351
+ "learning_rate": 1.2502075780032792e-05,
352
+ "loss": 1.3211,
353
+ "step": 235
354
+ },
355
+ {
356
+ "epoch": 0.48879837067209775,
357
+ "grad_norm": 0.3314637232326215,
358
+ "learning_rate": 1.2155701619244997e-05,
359
+ "loss": 1.3146,
360
+ "step": 240
361
+ },
362
+ {
363
+ "epoch": 0.4989816700610998,
364
+ "grad_norm": 0.33428121648270825,
365
+ "learning_rate": 1.1806592786063991e-05,
366
+ "loss": 1.2953,
367
+ "step": 245
368
+ },
369
+ {
370
+ "epoch": 0.5091649694501018,
371
+ "grad_norm": 0.33140728066789105,
372
+ "learning_rate": 1.1455192151747931e-05,
373
+ "loss": 1.3083,
374
+ "step": 250
375
+ },
376
+ {
377
+ "epoch": 0.5193482688391039,
378
+ "grad_norm": 0.34894619572153845,
379
+ "learning_rate": 1.1101945494880013e-05,
380
+ "loss": 1.3001,
381
+ "step": 255
382
+ },
383
+ {
384
+ "epoch": 0.5295315682281059,
385
+ "grad_norm": 0.2936066479920224,
386
+ "learning_rate": 1.0747300935864245e-05,
387
+ "loss": 1.299,
388
+ "step": 260
389
+ },
390
+ {
391
+ "epoch": 0.539714867617108,
392
+ "grad_norm": 0.34761064821495535,
393
+ "learning_rate": 1.0391708368450429e-05,
394
+ "loss": 1.2987,
395
+ "step": 265
396
+ },
397
+ {
398
+ "epoch": 0.5498981670061099,
399
+ "grad_norm": 0.32443077572814916,
400
+ "learning_rate": 1.0035618889009535e-05,
401
+ "loss": 1.3154,
402
+ "step": 270
403
+ },
404
+ {
405
+ "epoch": 0.560081466395112,
406
+ "grad_norm": 0.3573334196924316,
407
+ "learning_rate": 9.67948422428345e-06,
408
+ "loss": 1.3045,
409
+ "step": 275
410
+ },
411
+ {
412
+ "epoch": 0.570264765784114,
413
+ "grad_norm": 0.2929859629265363,
414
+ "learning_rate": 9.323756158335054e-06,
415
+ "loss": 1.2995,
416
+ "step": 280
417
+ },
418
+ {
419
+ "epoch": 0.5804480651731161,
420
+ "grad_norm": 0.3725484551984738,
421
+ "learning_rate": 8.968885959425567e-06,
422
+ "loss": 1.3006,
423
+ "step": 285
424
+ },
425
+ {
426
+ "epoch": 0.5906313645621182,
427
+ "grad_norm": 0.2977272122499357,
428
+ "learning_rate": 8.615323807546258e-06,
429
+ "loss": 1.2785,
430
+ "step": 290
431
+ },
432
+ {
433
+ "epoch": 0.6008146639511202,
434
+ "grad_norm": 0.2934843742212814,
435
+ "learning_rate": 8.263518223330698e-06,
436
+ "loss": 1.301,
437
+ "step": 295
438
+ },
439
+ {
440
+ "epoch": 0.6109979633401222,
441
+ "grad_norm": 0.2851617487482263,
442
+ "learning_rate": 7.913915499071994e-06,
443
+ "loss": 1.2957,
444
+ "step": 300
445
+ },
446
+ {
447
+ "epoch": 0.6211812627291242,
448
+ "grad_norm": 0.3280712460992494,
449
+ "learning_rate": 7.566959132566914e-06,
450
+ "loss": 1.3021,
451
+ "step": 305
452
+ },
453
+ {
454
+ "epoch": 0.6313645621181263,
455
+ "grad_norm": 0.28660010388812523,
456
+ "learning_rate": 7.223089264505001e-06,
457
+ "loss": 1.2809,
458
+ "step": 310
459
+ },
460
+ {
461
+ "epoch": 0.6415478615071283,
462
+ "grad_norm": 0.29055149959882093,
463
+ "learning_rate": 6.882742120116419e-06,
464
+ "loss": 1.2958,
465
+ "step": 315
466
+ },
467
+ {
468
+ "epoch": 0.6517311608961304,
469
+ "grad_norm": 0.2868148780704892,
470
+ "learning_rate": 6.546349455786926e-06,
471
+ "loss": 1.2931,
472
+ "step": 320
473
+ },
474
+ {
475
+ "epoch": 0.6619144602851323,
476
+ "grad_norm": 0.3021008995725944,
477
+ "learning_rate": 6.214338011341825e-06,
478
+ "loss": 1.2855,
479
+ "step": 325
480
+ },
481
+ {
482
+ "epoch": 0.6720977596741344,
483
+ "grad_norm": 0.2840488188778215,
484
+ "learning_rate": 5.887128968693887e-06,
485
+ "loss": 1.3091,
486
+ "step": 330
487
+ },
488
+ {
489
+ "epoch": 0.6822810590631364,
490
+ "grad_norm": 0.2978772611216691,
491
+ "learning_rate": 5.565137417541866e-06,
492
+ "loss": 1.2826,
493
+ "step": 335
494
+ },
495
+ {
496
+ "epoch": 0.6924643584521385,
497
+ "grad_norm": 0.28796550430490037,
498
+ "learning_rate": 5.248771828797474e-06,
499
+ "loss": 1.3008,
500
+ "step": 340
501
+ },
502
+ {
503
+ "epoch": 0.7026476578411406,
504
+ "grad_norm": 0.2791511099905026,
505
+ "learning_rate": 4.938433536408771e-06,
506
+ "loss": 1.2848,
507
+ "step": 345
508
+ },
509
+ {
510
+ "epoch": 0.7128309572301426,
511
+ "grad_norm": 0.282696156100298,
512
+ "learning_rate": 4.634516228237372e-06,
513
+ "loss": 1.291,
514
+ "step": 350
515
+ },
516
+ {
517
+ "epoch": 0.7230142566191446,
518
+ "grad_norm": 0.27007483942462107,
519
+ "learning_rate": 4.337405446635264e-06,
520
+ "loss": 1.2821,
521
+ "step": 355
522
+ },
523
+ {
524
+ "epoch": 0.7331975560081466,
525
+ "grad_norm": 0.2885536779707833,
526
+ "learning_rate": 4.047478099354857e-06,
527
+ "loss": 1.3193,
528
+ "step": 360
529
+ },
530
+ {
531
+ "epoch": 0.7433808553971487,
532
+ "grad_norm": 0.3194820894463116,
533
+ "learning_rate": 3.7651019814126656e-06,
534
+ "loss": 1.2993,
535
+ "step": 365
536
+ },
537
+ {
538
+ "epoch": 0.7535641547861507,
539
+ "grad_norm": 0.28463406586484574,
540
+ "learning_rate": 3.4906353085131917e-06,
541
+ "loss": 1.297,
542
+ "step": 370
543
+ },
544
+ {
545
+ "epoch": 0.7637474541751528,
546
+ "grad_norm": 0.2887353591273086,
547
+ "learning_rate": 3.224426262624908e-06,
548
+ "loss": 1.2887,
549
+ "step": 375
550
+ },
551
+ {
552
+ "epoch": 0.7739307535641547,
553
+ "grad_norm": 0.28334983310000655,
554
+ "learning_rate": 2.9668125502848035e-06,
555
+ "loss": 1.2951,
556
+ "step": 380
557
+ },
558
+ {
559
+ "epoch": 0.7841140529531568,
560
+ "grad_norm": 0.25257521474138683,
561
+ "learning_rate": 2.7181209741918093e-06,
562
+ "loss": 1.2937,
563
+ "step": 385
564
+ },
565
+ {
566
+ "epoch": 0.7942973523421588,
567
+ "grad_norm": 0.2871884940474984,
568
+ "learning_rate": 2.478667018632562e-06,
569
+ "loss": 1.2858,
570
+ "step": 390
571
+ },
572
+ {
573
+ "epoch": 0.8044806517311609,
574
+ "grad_norm": 0.25758574412430923,
575
+ "learning_rate": 2.2487544492654832e-06,
576
+ "loss": 1.3027,
577
+ "step": 395
578
+ },
579
+ {
580
+ "epoch": 0.814663951120163,
581
+ "grad_norm": 0.2840318008095242,
582
+ "learning_rate": 2.0286749277707783e-06,
583
+ "loss": 1.2961,
584
+ "step": 400
585
+ },
586
+ {
587
+ "epoch": 0.814663951120163,
588
+ "eval_loss": 1.297537922859192,
589
+ "eval_runtime": 58.3016,
590
+ "eval_samples_per_second": 238.536,
591
+ "eval_steps_per_second": 3.739,
592
+ "step": 400
593
+ },
594
+ {
595
+ "epoch": 0.824847250509165,
596
+ "grad_norm": 0.3101944505956513,
597
+ "learning_rate": 1.8187076418552974e-06,
598
+ "loss": 1.3014,
599
+ "step": 405
600
+ },
601
+ {
602
+ "epoch": 0.835030549898167,
603
+ "grad_norm": 0.2791695231780548,
604
+ "learning_rate": 1.6191189510815942e-06,
605
+ "loss": 1.2931,
606
+ "step": 410
607
+ },
608
+ {
609
+ "epoch": 0.845213849287169,
610
+ "grad_norm": 0.28415893155261507,
611
+ "learning_rate": 1.4301620489704072e-06,
612
+ "loss": 1.2988,
613
+ "step": 415
614
+ },
615
+ {
616
+ "epoch": 0.8553971486761711,
617
+ "grad_norm": 0.2822535369305096,
618
+ "learning_rate": 1.2520766418053408e-06,
619
+ "loss": 1.2905,
620
+ "step": 420
621
+ },
622
+ {
623
+ "epoch": 0.8655804480651731,
624
+ "grad_norm": 0.28168468277135167,
625
+ "learning_rate": 1.0850886445471055e-06,
626
+ "loss": 1.2817,
627
+ "step": 425
628
+ },
629
+ {
630
+ "epoch": 0.8757637474541752,
631
+ "grad_norm": 0.2655425105216882,
632
+ "learning_rate": 9.294098942430996e-07,
633
+ "loss": 1.2843,
634
+ "step": 430
635
+ },
636
+ {
637
+ "epoch": 0.8859470468431772,
638
+ "grad_norm": 0.2741436677332839,
639
+ "learning_rate": 7.852378812959227e-07,
640
+ "loss": 1.2924,
641
+ "step": 435
642
+ },
643
+ {
644
+ "epoch": 0.8961303462321792,
645
+ "grad_norm": 0.26895584158358093,
646
+ "learning_rate": 6.527554989316898e-07,
647
+ "loss": 1.2935,
648
+ "step": 440
649
+ },
650
+ {
651
+ "epoch": 0.9063136456211812,
652
+ "grad_norm": 0.266002518787797,
653
+ "learning_rate": 5.321308111859791e-07,
654
+ "loss": 1.2857,
655
+ "step": 445
656
+ },
657
+ {
658
+ "epoch": 0.9164969450101833,
659
+ "grad_norm": 0.2650609910106162,
660
+ "learning_rate": 4.235168397017542e-07,
661
+ "loss": 1.2664,
662
+ "step": 450
663
+ },
664
+ {
665
+ "epoch": 0.9266802443991853,
666
+ "grad_norm": 0.2647562856371479,
667
+ "learning_rate": 3.2705136960970554e-07,
668
+ "loss": 1.2988,
669
+ "step": 455
670
+ },
671
+ {
672
+ "epoch": 0.9368635437881874,
673
+ "grad_norm": 0.29894331322023837,
674
+ "learning_rate": 2.4285677473727123e-07,
675
+ "loss": 1.2744,
676
+ "step": 460
677
+ },
678
+ {
679
+ "epoch": 0.9470468431771895,
680
+ "grad_norm": 0.2719080220558317,
681
+ "learning_rate": 1.7103986236807312e-07,
682
+ "loss": 1.2803,
683
+ "step": 465
684
+ },
685
+ {
686
+ "epoch": 0.9572301425661914,
687
+ "grad_norm": 0.2791229463614563,
688
+ "learning_rate": 1.1169173774871478e-07,
689
+ "loss": 1.2906,
690
+ "step": 470
691
+ },
692
+ {
693
+ "epoch": 0.9674134419551935,
694
+ "grad_norm": 0.2987546290973262,
695
+ "learning_rate": 6.488768851480087e-08,
696
+ "loss": 1.2884,
697
+ "step": 475
698
+ },
699
+ {
700
+ "epoch": 0.9775967413441955,
701
+ "grad_norm": 0.30009885175708245,
702
+ "learning_rate": 3.0687089182819264e-08,
703
+ "loss": 1.3051,
704
+ "step": 480
705
+ },
706
+ {
707
+ "epoch": 0.9877800407331976,
708
+ "grad_norm": 0.2873283513110056,
709
+ "learning_rate": 9.13332582901716e-09,
710
+ "loss": 1.286,
711
+ "step": 485
712
+ },
713
+ {
714
+ "epoch": 0.9979633401221996,
715
+ "grad_norm": 0.272339006461075,
716
+ "learning_rate": 2.5374105085518297e-10,
717
+ "loss": 1.2873,
718
+ "step": 490
719
+ },
720
+ {
721
+ "epoch": 1.0,
722
+ "step": 491,
723
+ "total_flos": 154920007237632.0,
724
+ "train_loss": 1.4646558688760047,
725
+ "train_runtime": 2127.1747,
726
+ "train_samples_per_second": 59.07,
727
+ "train_steps_per_second": 0.231
728
+ }
729
+ ],
730
+ "logging_steps": 5,
731
+ "max_steps": 491,
732
+ "num_input_tokens_seen": 0,
733
+ "num_train_epochs": 1,
734
+ "save_steps": 1000000,
735
+ "stateful_callbacks": {
736
+ "TrainerControl": {
737
+ "args": {
738
+ "should_epoch_stop": false,
739
+ "should_evaluate": false,
740
+ "should_log": false,
741
+ "should_save": true,
742
+ "should_training_stop": true
743
+ },
744
+ "attributes": {}
745
+ }
746
+ },
747
+ "total_flos": 154920007237632.0,
748
+ "train_batch_size": 8,
749
+ "trial_name": null,
750
+ "trial_params": null
751
+ }