chansung commited on
Commit
40d236c
·
verified ·
1 Parent(s): aa36d55

Model save

Browse files
README.md ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: gemma
4
+ base_model: google/gemma-7b
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: gemma7b-gpt4o_1k_coding-fft
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
+ # gemma7b-gpt4o_1k_coding-fft
20
+
21
+ This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 4.8389
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: 0.0003
43
+ - train_batch_size: 2
44
+ - eval_batch_size: 2
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 8
48
+ - gradient_accumulation_steps: 2
49
+ - total_train_batch_size: 32
50
+ - total_eval_batch_size: 16
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
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
+ | 1.8 | 1.0 | 598 | 4.8389 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - Transformers 4.45.1
66
+ - Pytorch 2.4.1+cu121
67
+ - Datasets 3.0.1
68
+ - Tokenizers 0.20.0
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 20570041221120.0,
4
+ "train_loss": 5.0949278754932825,
5
+ "train_runtime": 8333.1384,
6
+ "train_samples": 116368,
7
+ "train_samples_per_second": 2.295,
8
+ "train_steps_per_second": 0.072
9
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 2,
4
+ "eos_token_id": 1,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.45.1"
7
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 20570041221120.0,
4
+ "train_loss": 5.0949278754932825,
5
+ "train_runtime": 8333.1384,
6
+ "train_samples": 116368,
7
+ "train_samples_per_second": 2.295,
8
+ "train_steps_per_second": 0.072
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,890 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 598,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0016722408026755853,
13
+ "grad_norm": 5899.944994453189,
14
+ "learning_rate": 4.9999999999999996e-06,
15
+ "loss": 22.9452,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.008361204013377926,
20
+ "grad_norm": 3832.174007639646,
21
+ "learning_rate": 2.4999999999999998e-05,
22
+ "loss": 17.2666,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.016722408026755852,
27
+ "grad_norm": 1842.7301746786366,
28
+ "learning_rate": 4.9999999999999996e-05,
29
+ "loss": 16.1655,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.02508361204013378,
34
+ "grad_norm": 342.882874624202,
35
+ "learning_rate": 7.5e-05,
36
+ "loss": 6.9604,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.033444816053511704,
41
+ "grad_norm": 648.6675239239157,
42
+ "learning_rate": 9.999999999999999e-05,
43
+ "loss": 5.362,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.04180602006688963,
48
+ "grad_norm": 642.7533463105359,
49
+ "learning_rate": 0.000125,
50
+ "loss": 7.6165,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.05016722408026756,
55
+ "grad_norm": 124.26540956463161,
56
+ "learning_rate": 0.00015,
57
+ "loss": 5.5364,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.05852842809364549,
62
+ "grad_norm": 527.1226263500417,
63
+ "learning_rate": 0.000175,
64
+ "loss": 6.2687,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.06688963210702341,
69
+ "grad_norm": 235.15362781581305,
70
+ "learning_rate": 0.00019999999999999998,
71
+ "loss": 6.8242,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.07525083612040134,
76
+ "grad_norm": 232.04999588659172,
77
+ "learning_rate": 0.000225,
78
+ "loss": 5.0975,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.08361204013377926,
83
+ "grad_norm": 193.92802731203975,
84
+ "learning_rate": 0.00025,
85
+ "loss": 4.363,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.09197324414715718,
90
+ "grad_norm": 87.32996989626787,
91
+ "learning_rate": 0.00027499999999999996,
92
+ "loss": 4.492,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.10033444816053512,
97
+ "grad_norm": 157.51219806377262,
98
+ "learning_rate": 0.0003,
99
+ "loss": 5.2962,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.10869565217391304,
104
+ "grad_norm": 178.03521535228302,
105
+ "learning_rate": 0.0002999360698660697,
106
+ "loss": 5.2349,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.11705685618729098,
111
+ "grad_norm": 82.65266770623266,
112
+ "learning_rate": 0.0002997443339584391,
113
+ "loss": 4.8057,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.1254180602006689,
118
+ "grad_norm": 48.21827434938864,
119
+ "learning_rate": 0.0002994249557131383,
120
+ "loss": 4.6267,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.13377926421404682,
125
+ "grad_norm": 101.47726512121915,
126
+ "learning_rate": 0.00029897820736875393,
127
+ "loss": 3.7307,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.14214046822742474,
132
+ "grad_norm": 84.76151041346198,
133
+ "learning_rate": 0.0002984044697343724,
134
+ "loss": 3.3221,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.1505016722408027,
139
+ "grad_norm": 35.2161154461866,
140
+ "learning_rate": 0.00029770423186497794,
141
+ "loss": 2.7633,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.1588628762541806,
146
+ "grad_norm": 39.923892471272296,
147
+ "learning_rate": 0.0002968780906445808,
148
+ "loss": 2.1568,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.16722408026755853,
153
+ "grad_norm": 72.82838678501577,
154
+ "learning_rate": 0.00029592675027743253,
155
+ "loss": 2.3427,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.17558528428093645,
160
+ "grad_norm": 94.37110002387033,
161
+ "learning_rate": 0.00029485102168776096,
162
+ "loss": 6.1606,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.18394648829431437,
167
+ "grad_norm": 373.81535831887,
168
+ "learning_rate": 0.0002936518218285368,
169
+ "loss": 10.4362,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.19230769230769232,
174
+ "grad_norm": 221.23048878415432,
175
+ "learning_rate": 0.0002923301728998616,
176
+ "loss": 16.4553,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.20066889632107024,
181
+ "grad_norm": 395.57153442034314,
182
+ "learning_rate": 0.0002908872014776423,
183
+ "loss": 43.516,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.20903010033444816,
188
+ "grad_norm": 286.615099772232,
189
+ "learning_rate": 0.0002893241375532959,
190
+ "loss": 28.7129,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.21739130434782608,
195
+ "grad_norm": 82.2777934481911,
196
+ "learning_rate": 0.00028764231348530267,
197
+ "loss": 27.8137,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.225752508361204,
202
+ "grad_norm": 170.02097895100943,
203
+ "learning_rate": 0.00028584316286350164,
204
+ "loss": 21.8417,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.23411371237458195,
209
+ "grad_norm": 202.04237746771835,
210
+ "learning_rate": 0.0002839282192870955,
211
+ "loss": 19.1217,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.24247491638795987,
216
+ "grad_norm": 34.74067140580764,
217
+ "learning_rate": 0.00028189911505740843,
218
+ "loss": 15.3816,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.2508361204013378,
223
+ "grad_norm": 59.32768633148153,
224
+ "learning_rate": 0.0002797575797865093,
225
+ "loss": 13.5017,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.2591973244147157,
230
+ "grad_norm": 45.03643469009911,
231
+ "learning_rate": 0.0002775054389228871,
232
+ "loss": 11.8208,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.26755852842809363,
237
+ "grad_norm": 29.78986842292205,
238
+ "learning_rate": 0.0002751446121954359,
239
+ "loss": 9.8787,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.27591973244147155,
244
+ "grad_norm": 19.68751651825531,
245
+ "learning_rate": 0.00027267711197707386,
246
+ "loss": 8.1994,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.2842809364548495,
251
+ "grad_norm": 16.564772134956407,
252
+ "learning_rate": 0.0002701050415693934,
253
+ "loss": 7.3116,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.29264214046822745,
258
+ "grad_norm": 18.01034438896602,
259
+ "learning_rate": 0.0002674305934098032,
260
+ "loss": 6.7163,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.3010033444816054,
265
+ "grad_norm": 12.65556929182177,
266
+ "learning_rate": 0.00026465604720269026,
267
+ "loss": 6.2962,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.3093645484949833,
272
+ "grad_norm": 9.795061211752262,
273
+ "learning_rate": 0.0002617837679761962,
274
+ "loss": 6.0851,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.3177257525083612,
279
+ "grad_norm": 9.27750296099571,
280
+ "learning_rate": 0.0002588162040662628,
281
+ "loss": 5.8121,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.32608695652173914,
286
+ "grad_norm": 7.581802949350765,
287
+ "learning_rate": 0.0002557558850296661,
288
+ "loss": 5.5399,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.33444816053511706,
293
+ "grad_norm": 10.289352311810807,
294
+ "learning_rate": 0.00025260541948781784,
295
+ "loss": 5.3443,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.342809364548495,
300
+ "grad_norm": 6.606792882694537,
301
+ "learning_rate": 0.00024936749290317196,
302
+ "loss": 5.0023,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.3511705685618729,
307
+ "grad_norm": 5.306167947282732,
308
+ "learning_rate": 0.00024604486529013106,
309
+ "loss": 4.8264,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.3595317725752508,
314
+ "grad_norm": 8.160432944309026,
315
+ "learning_rate": 0.00024264036886240603,
316
+ "loss": 4.8126,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.36789297658862874,
321
+ "grad_norm": 9.350662398086465,
322
+ "learning_rate": 0.00023915690561883125,
323
+ "loss": 4.5927,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.3762541806020067,
328
+ "grad_norm": 6.717782881890457,
329
+ "learning_rate": 0.00023559744486969622,
330
+ "loss": 4.3447,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.38461538461538464,
335
+ "grad_norm": 10.834440165150719,
336
+ "learning_rate": 0.00023196502070569963,
337
+ "loss": 4.7644,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.39297658862876256,
342
+ "grad_norm": 9.530980370244805,
343
+ "learning_rate": 0.00022826272941168556,
344
+ "loss": 4.5336,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.4013377926421405,
349
+ "grad_norm": 9.96662219085004,
350
+ "learning_rate": 0.00022449372682736428,
351
+ "loss": 4.1642,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.4096989966555184,
356
+ "grad_norm": 7.17474089656695,
357
+ "learning_rate": 0.00022066122565726914,
358
+ "loss": 4.1165,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.4180602006688963,
363
+ "grad_norm": 5.991132986370517,
364
+ "learning_rate": 0.00021676849273224132,
365
+ "loss": 3.9771,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.42642140468227424,
370
+ "grad_norm": 6.305690862174887,
371
+ "learning_rate": 0.00021281884622477757,
372
+ "loss": 3.8831,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.43478260869565216,
377
+ "grad_norm": 5.754061162264476,
378
+ "learning_rate": 0.00020881565282061386,
379
+ "loss": 3.7495,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.4431438127090301,
384
+ "grad_norm": 5.539202432443519,
385
+ "learning_rate": 0.00020476232484895653,
386
+ "loss": 3.5914,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.451505016722408,
391
+ "grad_norm": 6.726459928222644,
392
+ "learning_rate": 0.00020066231737380684,
393
+ "loss": 3.5781,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.459866220735786,
398
+ "grad_norm": 5.078063325340097,
399
+ "learning_rate": 0.00019651912524885814,
400
+ "loss": 3.5243,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.4682274247491639,
405
+ "grad_norm": 6.020230646512102,
406
+ "learning_rate": 0.00019233628013847636,
407
+ "loss": 3.386,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.4765886287625418,
412
+ "grad_norm": 4.216873427535516,
413
+ "learning_rate": 0.00018811734750730306,
414
+ "loss": 3.4131,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.48494983277591974,
419
+ "grad_norm": 4.224871355162302,
420
+ "learning_rate": 0.000183865923581047,
421
+ "loss": 3.2989,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.49331103678929766,
426
+ "grad_norm": 4.645722524256328,
427
+ "learning_rate": 0.00017958563228105473,
428
+ "loss": 3.227,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.5016722408026756,
433
+ "grad_norm": 4.526907081748694,
434
+ "learning_rate": 0.000175280122135274,
435
+ "loss": 3.1885,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.5100334448160535,
440
+ "grad_norm": 3.8079799771786558,
441
+ "learning_rate": 0.00017095306316824145,
442
+ "loss": 3.1247,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.5183946488294314,
447
+ "grad_norm": 3.7709095979542058,
448
+ "learning_rate": 0.00016660814377274764,
449
+ "loss": 3.1044,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.5267558528428093,
454
+ "grad_norm": 3.7946474549046267,
455
+ "learning_rate": 0.00016224906756584414,
456
+ "loss": 3.0063,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.5351170568561873,
461
+ "grad_norm": 4.245301792150929,
462
+ "learning_rate": 0.00015787955023187394,
463
+ "loss": 2.9188,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.5434782608695652,
468
+ "grad_norm": 4.017268668767216,
469
+ "learning_rate": 0.00015350331635521528,
470
+ "loss": 2.9589,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.5518394648829431,
475
+ "grad_norm": 3.333420461437936,
476
+ "learning_rate": 0.00014912409624543942,
477
+ "loss": 2.8643,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.560200668896321,
482
+ "grad_norm": 3.2798450150057508,
483
+ "learning_rate": 0.000144745622757588,
484
+ "loss": 2.7767,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.568561872909699,
489
+ "grad_norm": 3.664786491151442,
490
+ "learning_rate": 0.00014037162811028105,
491
+ "loss": 2.6968,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.5769230769230769,
496
+ "grad_norm": 3.0151240913774635,
497
+ "learning_rate": 0.00013600584070436663,
498
+ "loss": 2.7462,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.5852842809364549,
503
+ "grad_norm": 4.300668720491546,
504
+ "learning_rate": 0.0001316519819448257,
505
+ "loss": 2.6857,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.5936454849498328,
510
+ "grad_norm": 6.856565698153125,
511
+ "learning_rate": 0.0001273137630686397,
512
+ "loss": 2.6735,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.6020066889632107,
517
+ "grad_norm": 3.6559482252841904,
518
+ "learning_rate": 0.0001229948819813257,
519
+ "loss": 2.6167,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.6103678929765887,
524
+ "grad_norm": 3.549605686334841,
525
+ "learning_rate": 0.00011869902010483487,
526
+ "loss": 2.5546,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.6187290969899666,
531
+ "grad_norm": 3.008473775307977,
532
+ "learning_rate": 0.00011442983923950197,
533
+ "loss": 2.5618,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.6270903010033445,
538
+ "grad_norm": 3.493843480596192,
539
+ "learning_rate": 0.00011019097844272025,
540
+ "loss": 2.4395,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.6354515050167224,
545
+ "grad_norm": 3.022476308004379,
546
+ "learning_rate": 0.00010598605092700247,
547
+ "loss": 2.4327,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.6438127090301003,
552
+ "grad_norm": 3.033725108847574,
553
+ "learning_rate": 0.00010181864098007187,
554
+ "loss": 2.4019,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.6521739130434783,
559
+ "grad_norm": 2.730168253261192,
560
+ "learning_rate": 9.76923009096091e-05,
561
+ "loss": 2.3644,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.6605351170568562,
566
+ "grad_norm": 2.5487378528989986,
567
+ "learning_rate": 9.361054801525882e-05,
568
+ "loss": 2.3956,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.6688963210702341,
573
+ "grad_norm": 3.0219826290571214,
574
+ "learning_rate": 8.9576861590477e-05,
575
+ "loss": 2.3118,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.677257525083612,
580
+ "grad_norm": 2.3724800707991442,
581
+ "learning_rate": 8.559467995677535e-05,
582
+ "loss": 2.3294,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.68561872909699,
587
+ "grad_norm": 3.218242001430894,
588
+ "learning_rate": 8.166739753288954e-05,
589
+ "loss": 2.2735,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.6939799331103679,
594
+ "grad_norm": 2.618273238863685,
595
+ "learning_rate": 7.77983619413708e-05,
596
+ "loss": 2.2728,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.7023411371237458,
601
+ "grad_norm": 2.1838152570753016,
602
+ "learning_rate": 7.399087115506633e-05,
603
+ "loss": 2.2301,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.7107023411371237,
608
+ "grad_norm": 2.3293111793801473,
609
+ "learning_rate": 7.024817068592166e-05,
610
+ "loss": 2.2563,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.7190635451505016,
615
+ "grad_norm": 3.3354279232331505,
616
+ "learning_rate": 6.657345081850015e-05,
617
+ "loss": 2.2324,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.7274247491638796,
622
+ "grad_norm": 2.4637443882850474,
623
+ "learning_rate": 6.296984389057886e-05,
624
+ "loss": 2.1572,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.7357859531772575,
629
+ "grad_norm": 2.66480391294473,
630
+ "learning_rate": 5.944042162313818e-05,
631
+ "loss": 2.1485,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.7441471571906354,
636
+ "grad_norm": 2.369277046664397,
637
+ "learning_rate": 5.59881925020216e-05,
638
+ "loss": 2.1156,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.7525083612040134,
643
+ "grad_norm": 2.404749403147846,
644
+ "learning_rate": 5.2616099213496704e-05,
645
+ "loss": 2.0883,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.7608695652173914,
650
+ "grad_norm": 2.151600123558966,
651
+ "learning_rate": 4.93270161359043e-05,
652
+ "loss": 2.0871,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.7692307692307693,
657
+ "grad_norm": 1.8485834869843754,
658
+ "learning_rate": 4.612374688953315e-05,
659
+ "loss": 2.0717,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.7775919732441472,
664
+ "grad_norm": 2.015234182046728,
665
+ "learning_rate": 4.300902194680902e-05,
666
+ "loss": 2.0691,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.7859531772575251,
671
+ "grad_norm": 1.8364808711885607,
672
+ "learning_rate": 3.998549630483525e-05,
673
+ "loss": 2.0201,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.794314381270903,
678
+ "grad_norm": 1.9757575481092078,
679
+ "learning_rate": 3.7055747222268256e-05,
680
+ "loss": 2.0128,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.802675585284281,
685
+ "grad_norm": 1.957601461432362,
686
+ "learning_rate": 3.422227202245778e-05,
687
+ "loss": 1.975,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.8110367892976589,
692
+ "grad_norm": 1.8124541787764386,
693
+ "learning_rate": 3.1487485964723966e-05,
694
+ "loss": 1.9249,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.8193979933110368,
699
+ "grad_norm": 1.8489721306641316,
700
+ "learning_rate": 2.885372018558604e-05,
701
+ "loss": 1.9786,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.8277591973244147,
706
+ "grad_norm": 1.9783817862893105,
707
+ "learning_rate": 2.6323219711697368e-05,
708
+ "loss": 1.9287,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.8361204013377926,
713
+ "grad_norm": 1.9022882119693414,
714
+ "learning_rate": 2.3898141546180678e-05,
715
+ "loss": 1.9572,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.8444816053511706,
720
+ "grad_norm": 2.1784717203571673,
721
+ "learning_rate": 2.1580552829994785e-05,
722
+ "loss": 1.9073,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.8528428093645485,
727
+ "grad_norm": 1.7051803845257447,
728
+ "learning_rate": 1.9372429079900052e-05,
729
+ "loss": 1.9092,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.8612040133779264,
734
+ "grad_norm": 1.8732499048540139,
735
+ "learning_rate": 1.727565250452414e-05,
736
+ "loss": 1.8558,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.8695652173913043,
741
+ "grad_norm": 1.4473569152727979,
742
+ "learning_rate": 1.52920103999642e-05,
743
+ "loss": 1.8463,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.8779264214046822,
748
+ "grad_norm": 1.4559349207330088,
749
+ "learning_rate": 1.3423193626292405e-05,
750
+ "loss": 1.8972,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.8862876254180602,
755
+ "grad_norm": 1.456537147004704,
756
+ "learning_rate": 1.1670795166263846e-05,
757
+ "loss": 1.8839,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.8946488294314381,
762
+ "grad_norm": 1.4711682274882243,
763
+ "learning_rate": 1.003630876745522e-05,
764
+ "loss": 1.8899,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.903010033444816,
769
+ "grad_norm": 1.3419796942254214,
770
+ "learning_rate": 8.521127668991612e-06,
771
+ "loss": 1.822,
772
+ "step": 540
773
+ },
774
+ {
775
+ "epoch": 0.9113712374581939,
776
+ "grad_norm": 1.4154994584871663,
777
+ "learning_rate": 7.126543413947045e-06,
778
+ "loss": 1.8409,
779
+ "step": 545
780
+ },
781
+ {
782
+ "epoch": 0.919732441471572,
783
+ "grad_norm": 1.5056657282270582,
784
+ "learning_rate": 5.853744748430883e-06,
785
+ "loss": 1.8138,
786
+ "step": 550
787
+ },
788
+ {
789
+ "epoch": 0.9280936454849499,
790
+ "grad_norm": 1.3367881313162435,
791
+ "learning_rate": 4.703816608298533e-06,
792
+ "loss": 1.8748,
793
+ "step": 555
794
+ },
795
+ {
796
+ "epoch": 0.9364548494983278,
797
+ "grad_norm": 1.4438743199653445,
798
+ "learning_rate": 3.6777391943500877e-06,
799
+ "loss": 1.8279,
800
+ "step": 560
801
+ },
802
+ {
803
+ "epoch": 0.9448160535117057,
804
+ "grad_norm": 1.3948588971264737,
805
+ "learning_rate": 2.7763871368055137e-06,
806
+ "loss": 1.8959,
807
+ "step": 565
808
+ },
809
+ {
810
+ "epoch": 0.9531772575250836,
811
+ "grad_norm": 1.2272041835054792,
812
+ "learning_rate": 2.000528749768199e-06,
813
+ "loss": 1.8279,
814
+ "step": 570
815
+ },
816
+ {
817
+ "epoch": 0.9615384615384616,
818
+ "grad_norm": 1.2479607608150853,
819
+ "learning_rate": 1.3508253763127974e-06,
820
+ "loss": 1.8121,
821
+ "step": 575
822
+ },
823
+ {
824
+ "epoch": 0.9698996655518395,
825
+ "grad_norm": 1.387198687782991,
826
+ "learning_rate": 8.278308247549992e-07,
827
+ "loss": 1.8638,
828
+ "step": 580
829
+ },
830
+ {
831
+ "epoch": 0.9782608695652174,
832
+ "grad_norm": 1.2432235153355073,
833
+ "learning_rate": 4.3199089658449425e-07,
834
+ "loss": 1.8267,
835
+ "step": 585
836
+ },
837
+ {
838
+ "epoch": 0.9866220735785953,
839
+ "grad_norm": 1.3076524438940778,
840
+ "learning_rate": 1.6364300646291018e-07,
841
+ "loss": 1.7537,
842
+ "step": 590
843
+ },
844
+ {
845
+ "epoch": 0.9949832775919732,
846
+ "grad_norm": 1.278244790033875,
847
+ "learning_rate": 2.30158946110115e-08,
848
+ "loss": 1.8,
849
+ "step": 595
850
+ },
851
+ {
852
+ "epoch": 1.0,
853
+ "eval_loss": 4.838874340057373,
854
+ "eval_runtime": 1.4359,
855
+ "eval_samples_per_second": 3.482,
856
+ "eval_steps_per_second": 0.696,
857
+ "step": 598
858
+ },
859
+ {
860
+ "epoch": 1.0,
861
+ "step": 598,
862
+ "total_flos": 20570041221120.0,
863
+ "train_loss": 5.0949278754932825,
864
+ "train_runtime": 8333.1384,
865
+ "train_samples_per_second": 2.295,
866
+ "train_steps_per_second": 0.072
867
+ }
868
+ ],
869
+ "logging_steps": 5,
870
+ "max_steps": 598,
871
+ "num_input_tokens_seen": 0,
872
+ "num_train_epochs": 1,
873
+ "save_steps": 100,
874
+ "stateful_callbacks": {
875
+ "TrainerControl": {
876
+ "args": {
877
+ "should_epoch_stop": false,
878
+ "should_evaluate": false,
879
+ "should_log": false,
880
+ "should_save": true,
881
+ "should_training_stop": true
882
+ },
883
+ "attributes": {}
884
+ }
885
+ },
886
+ "total_flos": 20570041221120.0,
887
+ "train_batch_size": 2,
888
+ "trial_name": null,
889
+ "trial_params": null
890
+ }