chansung commited on
Commit
c3c5c3b
1 Parent(s): 34a86e1

Model save

Browse files
README.md ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: llama3
4
+ base_model: meta-llama/Meta-Llama-3-8B
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: llama3.1-8b-gpt4o_100k_closedqa-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
+ # llama3.1-8b-gpt4o_100k_closedqa-fft
20
+
21
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 3.5964
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
+ | 0.9303 | 0.9990 | 511 | 3.5964 |
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": 0.9990224828934506,
3
+ "total_flos": 26722078556160.0,
4
+ "train_loss": 1.3314139091805235,
5
+ "train_runtime": 8881.3456,
6
+ "train_samples": 111440,
7
+ "train_samples_per_second": 1.842,
8
+ "train_steps_per_second": 0.058
9
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 128256,
3
+ "do_sample": true,
4
+ "eos_token_id": 128257,
5
+ "max_length": 4096,
6
+ "pad_token_id": 128257,
7
+ "temperature": 0.6,
8
+ "top_p": 0.9,
9
+ "transformers_version": "4.45.1"
10
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9990224828934506,
3
+ "total_flos": 26722078556160.0,
4
+ "train_loss": 1.3314139091805235,
5
+ "train_runtime": 8881.3456,
6
+ "train_samples": 111440,
7
+ "train_samples_per_second": 1.842,
8
+ "train_steps_per_second": 0.058
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,771 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9990224828934506,
5
+ "eval_steps": 500,
6
+ "global_step": 511,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0019550342130987292,
13
+ "grad_norm": 28.436507384573893,
14
+ "learning_rate": 5.769230769230769e-06,
15
+ "loss": 1.9755,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.009775171065493646,
20
+ "grad_norm": 16.326267191478145,
21
+ "learning_rate": 2.8846153846153845e-05,
22
+ "loss": 1.7228,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.019550342130987292,
27
+ "grad_norm": 7.77688805789614,
28
+ "learning_rate": 5.769230769230769e-05,
29
+ "loss": 1.1824,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.02932551319648094,
34
+ "grad_norm": 23.182562040558295,
35
+ "learning_rate": 8.653846153846152e-05,
36
+ "loss": 1.2132,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.039100684261974585,
41
+ "grad_norm": 3.8046090746822716,
42
+ "learning_rate": 0.00011538461538461538,
43
+ "loss": 1.0978,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.04887585532746823,
48
+ "grad_norm": 6.364935901579011,
49
+ "learning_rate": 0.00014423076923076922,
50
+ "loss": 1.2143,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.05865102639296188,
55
+ "grad_norm": 4.553589249592707,
56
+ "learning_rate": 0.00017307692307692304,
57
+ "loss": 1.1357,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.06842619745845552,
62
+ "grad_norm": 5.819131618520497,
63
+ "learning_rate": 0.00020192307692307691,
64
+ "loss": 1.2003,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.07820136852394917,
69
+ "grad_norm": 8.631833955662083,
70
+ "learning_rate": 0.00023076923076923076,
71
+ "loss": 1.2049,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.08797653958944282,
76
+ "grad_norm": 27.119527389294134,
77
+ "learning_rate": 0.0002596153846153846,
78
+ "loss": 2.0482,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.09775171065493646,
83
+ "grad_norm": 32.33385091368177,
84
+ "learning_rate": 0.00028846153846153843,
85
+ "loss": 1.6455,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.10752688172043011,
90
+ "grad_norm": 45.94902359405063,
91
+ "learning_rate": 0.0002999683799255387,
92
+ "loss": 1.9763,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.11730205278592376,
97
+ "grad_norm": 128.59363629595714,
98
+ "learning_rate": 0.0002997751944121241,
99
+ "loss": 1.5422,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.1270772238514174,
104
+ "grad_norm": 10.71651832916556,
105
+ "learning_rate": 0.0002994066160471166,
106
+ "loss": 1.7548,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.13685239491691104,
111
+ "grad_norm": 4.189410908699583,
112
+ "learning_rate": 0.0002988630764507904,
113
+ "loss": 1.3404,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.1466275659824047,
118
+ "grad_norm": 33.78267011030123,
119
+ "learning_rate": 0.00029814521213014585,
120
+ "loss": 1.4341,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.15640273704789834,
125
+ "grad_norm": 4.447241036402298,
126
+ "learning_rate": 0.00029725386373353455,
127
+ "loss": 1.4355,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.16617790811339198,
132
+ "grad_norm": 19.324376604180756,
133
+ "learning_rate": 0.00029619007506622504,
134
+ "loss": 1.4037,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.17595307917888564,
139
+ "grad_norm": 56.398325865658165,
140
+ "learning_rate": 0.00029495509186806487,
141
+ "loss": 2.1883,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.18572825024437928,
146
+ "grad_norm": 117.39758059322921,
147
+ "learning_rate": 0.0002935503603546683,
148
+ "loss": 2.0507,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.19550342130987292,
153
+ "grad_norm": 7.6884925370685995,
154
+ "learning_rate": 0.00029197752552383914,
155
+ "loss": 1.5932,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.20527859237536658,
160
+ "grad_norm": 963.0390862127407,
161
+ "learning_rate": 0.000290238429229211,
162
+ "loss": 5.223,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.21505376344086022,
167
+ "grad_norm": 258.2156128276988,
168
+ "learning_rate": 0.00028833510802336203,
169
+ "loss": 4.5213,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.22482893450635386,
174
+ "grad_norm": 264.762100287562,
175
+ "learning_rate": 0.0002862697907729285,
176
+ "loss": 2.0849,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.23460410557184752,
181
+ "grad_norm": 18.694604887076736,
182
+ "learning_rate": 0.0002840448960485118,
183
+ "loss": 1.8192,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.24437927663734116,
188
+ "grad_norm": 67.0812503804155,
189
+ "learning_rate": 0.00028166302929243326,
190
+ "loss": 1.3915,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.2541544477028348,
195
+ "grad_norm": 21.92554603766507,
196
+ "learning_rate": 0.0002791269797676551,
197
+ "loss": 1.5317,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.26392961876832843,
202
+ "grad_norm": 7.078011892011075,
203
+ "learning_rate": 0.00027643971729144056,
204
+ "loss": 1.4673,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.27370478983382207,
209
+ "grad_norm": 9.631691414307214,
210
+ "learning_rate": 0.0002736043887575761,
211
+ "loss": 1.3131,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.28347996089931576,
216
+ "grad_norm": 23.096760351073254,
217
+ "learning_rate": 0.00027062431445123124,
218
+ "loss": 1.572,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.2932551319648094,
223
+ "grad_norm": 3.17350344043114,
224
+ "learning_rate": 0.0002675029841607691,
225
+ "loss": 1.3668,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.30303030303030304,
230
+ "grad_norm": 6.678371795997961,
231
+ "learning_rate": 0.00026424405309106216,
232
+ "loss": 1.3082,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.3128054740957967,
237
+ "grad_norm": 4.718771263163467,
238
+ "learning_rate": 0.00026085133758309883,
239
+ "loss": 1.3286,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.3225806451612903,
244
+ "grad_norm": 8.790382644648172,
245
+ "learning_rate": 0.00025732881064489233,
246
+ "loss": 1.3241,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.33235581622678395,
251
+ "grad_norm": 4.416208770248871,
252
+ "learning_rate": 0.0002536805972989267,
253
+ "loss": 1.3357,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.3421309872922776,
258
+ "grad_norm": 3.1002678359027285,
259
+ "learning_rate": 0.0002499109697515875,
260
+ "loss": 1.4037,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.3519061583577713,
265
+ "grad_norm": 7.124469693986668,
266
+ "learning_rate": 0.0002460243423902342,
267
+ "loss": 1.625,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.3616813294232649,
272
+ "grad_norm": 51.006744297270856,
273
+ "learning_rate": 0.00024202526661377277,
274
+ "loss": 1.6499,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.37145650048875856,
279
+ "grad_norm": 14.662023479458805,
280
+ "learning_rate": 0.00023791842550278217,
281
+ "loss": 1.8342,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.3812316715542522,
286
+ "grad_norm": 4.973242239066626,
287
+ "learning_rate": 0.00023370862833543648,
288
+ "loss": 1.6823,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.39100684261974583,
293
+ "grad_norm": 95.12537291599145,
294
+ "learning_rate": 0.0002294008049556441,
295
+ "loss": 1.5268,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.40078201368523947,
300
+ "grad_norm": 3.394429345078054,
301
+ "learning_rate": 0.000225,
302
+ "loss": 1.45,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.41055718475073316,
307
+ "grad_norm": 3.5902349674560456,
308
+ "learning_rate": 0.00022051136699031057,
309
+ "loss": 1.2502,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.4203323558162268,
314
+ "grad_norm": 260.2217345574012,
315
+ "learning_rate": 0.00021594016229861007,
316
+ "loss": 1.4486,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.43010752688172044,
321
+ "grad_norm": 4.8647475236367725,
322
+ "learning_rate": 0.0002112917389917347,
323
+ "loss": 1.486,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.4398826979472141,
328
+ "grad_norm": 1.9024756731821182,
329
+ "learning_rate": 0.0002065715405626634,
330
+ "loss": 1.2628,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.4496578690127077,
335
+ "grad_norm": 10.37907922884646,
336
+ "learning_rate": 0.00020178509455596596,
337
+ "loss": 1.2518,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.45943304007820135,
342
+ "grad_norm": 1.9405506982628546,
343
+ "learning_rate": 0.00019693800609482315,
344
+ "loss": 1.2849,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.46920821114369504,
349
+ "grad_norm": 2.6160283264932462,
350
+ "learning_rate": 0.00019203595131719932,
351
+ "loss": 1.2548,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.4789833822091887,
356
+ "grad_norm": 2.1695347772705373,
357
+ "learning_rate": 0.00018708467072885382,
358
+ "loss": 1.3377,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.4887585532746823,
363
+ "grad_norm": 2.2144620011763374,
364
+ "learning_rate": 0.00018208996248097458,
365
+ "loss": 1.3093,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.49853372434017595,
370
+ "grad_norm": 2.1880901805448403,
371
+ "learning_rate": 0.00017705767558030754,
372
+ "loss": 1.245,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.5083088954056696,
377
+ "grad_norm": 2.991429961153096,
378
+ "learning_rate": 0.0001719937030397311,
379
+ "loss": 1.2559,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.5180840664711632,
384
+ "grad_norm": 23.064953881729117,
385
+ "learning_rate": 0.00016690397497729818,
386
+ "loss": 1.288,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.5278592375366569,
391
+ "grad_norm": 1.3455036549144204,
392
+ "learning_rate": 0.00016179445167182677,
393
+ "loss": 1.2717,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.5376344086021505,
398
+ "grad_norm": 1.1846238387921606,
399
+ "learning_rate": 0.00015667111658317054,
400
+ "loss": 1.2394,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.5474095796676441,
405
+ "grad_norm": 137.50996798714343,
406
+ "learning_rate": 0.00015153996934534348,
407
+ "loss": 1.3296,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.5571847507331378,
412
+ "grad_norm": 1.1488279921928382,
413
+ "learning_rate": 0.00014640701874070455,
414
+ "loss": 1.2874,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.5669599217986315,
419
+ "grad_norm": 1.2689991198054311,
420
+ "learning_rate": 0.00014127827566342863,
421
+ "loss": 1.2561,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.5767350928641252,
426
+ "grad_norm": 2.140511295392104,
427
+ "learning_rate": 0.0001361597460805047,
428
+ "loss": 1.2205,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.5865102639296188,
433
+ "grad_norm": 12.286388401977892,
434
+ "learning_rate": 0.000131057423998504,
435
+ "loss": 1.252,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.5962854349951124,
440
+ "grad_norm": 2.5541457654289395,
441
+ "learning_rate": 0.00012597728444435418,
442
+ "loss": 1.215,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.6060606060606061,
447
+ "grad_norm": 1.1732833800621696,
448
+ "learning_rate": 0.00012092527646833949,
449
+ "loss": 1.2053,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.6158357771260997,
454
+ "grad_norm": 1.4481798374657,
455
+ "learning_rate": 0.00011590731617752066,
456
+ "loss": 1.2061,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.6256109481915934,
461
+ "grad_norm": 0.9912604590459435,
462
+ "learning_rate": 0.00011092927980773267,
463
+ "loss": 1.1604,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.635386119257087,
468
+ "grad_norm": 1.0322599469502478,
469
+ "learning_rate": 0.00010599699684227311,
470
+ "loss": 1.1369,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.6451612903225806,
475
+ "grad_norm": 1.5437893851108073,
476
+ "learning_rate": 0.00010111624318534006,
477
+ "loss": 1.1721,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.6549364613880743,
482
+ "grad_norm": 1.383693940282455,
483
+ "learning_rate": 9.629273439821313e-05,
484
+ "loss": 1.1094,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.6647116324535679,
489
+ "grad_norm": 2.598856235735631,
490
+ "learning_rate": 9.15321190060981e-05,
491
+ "loss": 1.1251,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.6744868035190615,
496
+ "grad_norm": 1.0440965009342513,
497
+ "learning_rate": 8.683997188347435e-05,
498
+ "loss": 1.0953,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.6842619745845552,
503
+ "grad_norm": 1.2172823916521676,
504
+ "learning_rate": 8.222178772568959e-05,
505
+ "loss": 1.0839,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.6940371456500489,
510
+ "grad_norm": 0.9291103607651107,
511
+ "learning_rate": 7.768297461444765e-05,
512
+ "loss": 1.0786,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.7038123167155426,
517
+ "grad_norm": 36.656353739168324,
518
+ "learning_rate": 7.32288476847252e-05,
519
+ "loss": 1.1001,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.7135874877810362,
524
+ "grad_norm": 0.7923067460462172,
525
+ "learning_rate": 6.886462290053158e-05,
526
+ "loss": 1.0793,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.7233626588465298,
531
+ "grad_norm": 0.8419278909431203,
532
+ "learning_rate": 6.4595410946803e-05,
533
+ "loss": 1.0869,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.7331378299120235,
538
+ "grad_norm": 1.0339571657214093,
539
+ "learning_rate": 6.04262112445821e-05,
540
+ "loss": 1.0128,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.7429130009775171,
545
+ "grad_norm": 0.7333799573780848,
546
+ "learning_rate": 5.636190609649249e-05,
547
+ "loss": 1.0101,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.7526881720430108,
552
+ "grad_norm": 0.8595033406539786,
553
+ "learning_rate": 5.240725496936372e-05,
554
+ "loss": 1.0224,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.7624633431085044,
559
+ "grad_norm": 0.7112388580251547,
560
+ "learning_rate": 4.8566888920701196e-05,
561
+ "loss": 1.0016,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.772238514173998,
566
+ "grad_norm": 0.8167634152987004,
567
+ "learning_rate": 4.48453051755301e-05,
568
+ "loss": 0.9793,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.7820136852394917,
573
+ "grad_norm": 0.6682157986400304,
574
+ "learning_rate": 4.12468618599611e-05,
575
+ "loss": 1.0015,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.7917888563049853,
580
+ "grad_norm": 0.7797058649722416,
581
+ "learning_rate": 3.777577289764752e-05,
582
+ "loss": 0.9784,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.8015640273704789,
587
+ "grad_norm": 0.6768885831261553,
588
+ "learning_rate": 3.443610307510907e-05,
589
+ "loss": 0.9605,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.8113391984359726,
594
+ "grad_norm": 0.6416162504789994,
595
+ "learning_rate": 3.1231763281701305e-05,
596
+ "loss": 0.971,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.8211143695014663,
601
+ "grad_norm": 0.6784950124595185,
602
+ "learning_rate": 2.816650592980495e-05,
603
+ "loss": 0.9553,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.83088954056696,
608
+ "grad_norm": 0.689853020596839,
609
+ "learning_rate": 2.5243920560598184e-05,
610
+ "loss": 0.9351,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.8406647116324536,
615
+ "grad_norm": 0.6460996639595076,
616
+ "learning_rate": 2.24674296405579e-05,
617
+ "loss": 0.9313,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.8504398826979472,
622
+ "grad_norm": 0.6636695831614443,
623
+ "learning_rate": 1.98402845536117e-05,
624
+ "loss": 0.9266,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.8602150537634409,
629
+ "grad_norm": 0.6119153060968766,
630
+ "learning_rate": 1.736556179363543e-05,
631
+ "loss": 0.9134,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.8699902248289345,
636
+ "grad_norm": 0.5955119176542452,
637
+ "learning_rate": 1.5046159361753224e-05,
638
+ "loss": 0.9198,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.8797653958944281,
643
+ "grad_norm": 0.6324852203366718,
644
+ "learning_rate": 1.2884793372660207e-05,
645
+ "loss": 0.9051,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.8895405669599218,
650
+ "grad_norm": 0.6803072103352026,
651
+ "learning_rate": 1.0883994873941815e-05,
652
+ "loss": 0.8923,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.8993157380254154,
657
+ "grad_norm": 2.125025596226289,
658
+ "learning_rate": 9.046106882113751e-06,
659
+ "loss": 0.9315,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.9090909090909091,
664
+ "grad_norm": 0.5812394213670566,
665
+ "learning_rate": 7.373281638854328e-06,
666
+ "loss": 0.9131,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.9188660801564027,
671
+ "grad_norm": 0.6819326019234143,
672
+ "learning_rate": 5.867478090641892e-06,
673
+ "loss": 0.9521,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.9286412512218963,
678
+ "grad_norm": 0.6009777818830081,
679
+ "learning_rate": 4.530459594748592e-06,
680
+ "loss": 0.8585,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.9384164222873901,
685
+ "grad_norm": 0.5996940109383458,
686
+ "learning_rate": 3.363791854277348e-06,
687
+ "loss": 0.8938,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.9481915933528837,
692
+ "grad_norm": 0.6175664551394122,
693
+ "learning_rate": 2.3688410846596282e-06,
694
+ "loss": 0.8891,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.9579667644183774,
699
+ "grad_norm": 2.0911762607681768,
700
+ "learning_rate": 1.5467724137617043e-06,
701
+ "loss": 0.924,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.967741935483871,
706
+ "grad_norm": 0.6696352577624144,
707
+ "learning_rate": 8.985485174722973e-07,
708
+ "loss": 0.9077,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.9775171065493646,
713
+ "grad_norm": 0.5766498486118373,
714
+ "learning_rate": 4.249284923700358e-07,
715
+ "loss": 0.9012,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.9872922776148583,
720
+ "grad_norm": 0.6254945643083392,
721
+ "learning_rate": 1.2646696679042833e-07,
722
+ "loss": 0.9035,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.9970674486803519,
727
+ "grad_norm": 0.6202670688713676,
728
+ "learning_rate": 3.5134513334200697e-09,
729
+ "loss": 0.9303,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.9990224828934506,
734
+ "eval_loss": 3.596351385116577,
735
+ "eval_runtime": 2.2495,
736
+ "eval_samples_per_second": 2.667,
737
+ "eval_steps_per_second": 0.445,
738
+ "step": 511
739
+ },
740
+ {
741
+ "epoch": 0.9990224828934506,
742
+ "step": 511,
743
+ "total_flos": 26722078556160.0,
744
+ "train_loss": 1.3314139091805235,
745
+ "train_runtime": 8881.3456,
746
+ "train_samples_per_second": 1.842,
747
+ "train_steps_per_second": 0.058
748
+ }
749
+ ],
750
+ "logging_steps": 5,
751
+ "max_steps": 511,
752
+ "num_input_tokens_seen": 0,
753
+ "num_train_epochs": 1,
754
+ "save_steps": 100,
755
+ "stateful_callbacks": {
756
+ "TrainerControl": {
757
+ "args": {
758
+ "should_epoch_stop": false,
759
+ "should_evaluate": false,
760
+ "should_log": false,
761
+ "should_save": true,
762
+ "should_training_stop": true
763
+ },
764
+ "attributes": {}
765
+ }
766
+ },
767
+ "total_flos": 26722078556160.0,
768
+ "train_batch_size": 2,
769
+ "trial_name": null,
770
+ "trial_params": null
771
+ }