zaas12 commited on
Commit
094217c
·
verified ·
1 Parent(s): 603feea

Upload folder using huggingface_hub

Browse files
.ipynb_checkpoints/config-checkpoint.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/phi-1_5",
3
+ "architectures": [
4
+ "PhiForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": null,
8
+ "embd_pdrop": 0.0,
9
+ "eos_token_id": null,
10
+ "hidden_act": "gelu_new",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8192,
14
+ "layer_norm_eps": 1e-05,
15
+ "max_position_embeddings": 2048,
16
+ "model_type": "phi",
17
+ "num_attention_heads": 32,
18
+ "num_hidden_layers": 24,
19
+ "num_key_value_heads": 32,
20
+ "partial_rotary_factor": 0.5,
21
+ "qk_layernorm": false,
22
+ "resid_pdrop": 0.0,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000.0,
25
+ "tie_word_embeddings": false,
26
+ "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.45.0",
28
+ "use_cache": false,
29
+ "vocab_size": 51200
30
+ }
.ipynb_checkpoints/generation_config-checkpoint.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.45.0"
4
+ }
checkpoint-2994/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/phi-1_5",
3
+ "architectures": [
4
+ "PhiForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": null,
8
+ "embd_pdrop": 0.0,
9
+ "eos_token_id": null,
10
+ "hidden_act": "gelu_new",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8192,
14
+ "layer_norm_eps": 1e-05,
15
+ "max_position_embeddings": 2048,
16
+ "model_type": "phi",
17
+ "num_attention_heads": 32,
18
+ "num_hidden_layers": 24,
19
+ "num_key_value_heads": 32,
20
+ "partial_rotary_factor": 0.5,
21
+ "qk_layernorm": false,
22
+ "resid_pdrop": 0.0,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000.0,
25
+ "tie_word_embeddings": false,
26
+ "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.45.0",
28
+ "use_cache": false,
29
+ "vocab_size": 51200
30
+ }
checkpoint-2994/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.45.0"
4
+ }
checkpoint-2994/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28afc072c0368ffb1e5a528422ef98a4d7bcfde48daeb85ef4c70b6e3fdab506
3
+ size 2836579040
checkpoint-2994/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fcfee381d271b30f63fb2353bd5caf11a7faf709a92f7f1d1d00426e4c2a09a6
3
+ size 11346377382
checkpoint-2994/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
3
+ size 14244
checkpoint-2994/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef73212274ea360a47c107bc0721c878dc6e4db3a7d41cc16c2ace3fbf6a7f74
3
+ size 1064
checkpoint-2994/trainer_state.json ADDED
@@ -0,0 +1,454 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.945128321647644,
3
+ "best_model_checkpoint": "./checkpoints/ultrafeedback_binarized/phi-1_5-ultrafeedback_binarized-lambda0.25-ORPO-18-16-19/checkpoint-2994",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2994,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.016700066800267203,
13
+ "grad_norm": 5.34375,
14
+ "learning_rate": 2.0000000000000002e-07,
15
+ "loss": 1.2526,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.033400133600534405,
20
+ "grad_norm": 3.453125,
21
+ "learning_rate": 4.0000000000000003e-07,
22
+ "loss": 1.2115,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.050100200400801605,
27
+ "grad_norm": 2.953125,
28
+ "learning_rate": 6.000000000000001e-07,
29
+ "loss": 1.2056,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.06680026720106881,
34
+ "grad_norm": 2.3125,
35
+ "learning_rate": 8.000000000000001e-07,
36
+ "loss": 1.2425,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.08350033400133601,
41
+ "grad_norm": 1.9921875,
42
+ "learning_rate": 1.0000000000000002e-06,
43
+ "loss": 1.2355,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.10020040080160321,
48
+ "grad_norm": 2.484375,
49
+ "learning_rate": 1.2000000000000002e-06,
50
+ "loss": 1.2408,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.11690046760187041,
55
+ "grad_norm": 2.453125,
56
+ "learning_rate": 1.4000000000000001e-06,
57
+ "loss": 1.2136,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.13360053440213762,
62
+ "grad_norm": 2.328125,
63
+ "learning_rate": 1.6000000000000001e-06,
64
+ "loss": 1.2241,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.15030060120240482,
69
+ "grad_norm": 3.859375,
70
+ "learning_rate": 1.8000000000000001e-06,
71
+ "loss": 1.213,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.16700066800267202,
76
+ "grad_norm": 2.90625,
77
+ "learning_rate": 2.0000000000000003e-06,
78
+ "loss": 1.2212,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.18370073480293922,
83
+ "grad_norm": 1.828125,
84
+ "learning_rate": 2.2e-06,
85
+ "loss": 1.1722,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.20040080160320642,
90
+ "grad_norm": 2.953125,
91
+ "learning_rate": 2.4000000000000003e-06,
92
+ "loss": 1.1541,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.21710086840347362,
97
+ "grad_norm": 4.3125,
98
+ "learning_rate": 2.6e-06,
99
+ "loss": 1.1767,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.23380093520374082,
104
+ "grad_norm": 2.828125,
105
+ "learning_rate": 2.8000000000000003e-06,
106
+ "loss": 1.0593,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.250501002004008,
111
+ "grad_norm": 1.7421875,
112
+ "learning_rate": 3e-06,
113
+ "loss": 1.128,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.26720106880427524,
118
+ "grad_norm": 2.34375,
119
+ "learning_rate": 3.2000000000000003e-06,
120
+ "loss": 1.1307,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.28390113560454244,
125
+ "grad_norm": 2.390625,
126
+ "learning_rate": 3.4000000000000005e-06,
127
+ "loss": 1.0742,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.30060120240480964,
132
+ "grad_norm": 1.921875,
133
+ "learning_rate": 3.6000000000000003e-06,
134
+ "loss": 1.078,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.31730126920507684,
139
+ "grad_norm": 2.4375,
140
+ "learning_rate": 3.8000000000000005e-06,
141
+ "loss": 1.063,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.33400133600534404,
146
+ "grad_norm": 2.59375,
147
+ "learning_rate": 4.000000000000001e-06,
148
+ "loss": 1.0211,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.35070140280561124,
153
+ "grad_norm": 1.7890625,
154
+ "learning_rate": 4.2000000000000004e-06,
155
+ "loss": 1.0235,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.36740146960587844,
160
+ "grad_norm": 2.46875,
161
+ "learning_rate": 4.4e-06,
162
+ "loss": 1.0346,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.38410153640614564,
167
+ "grad_norm": 1.0625,
168
+ "learning_rate": 4.600000000000001e-06,
169
+ "loss": 1.038,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.40080160320641284,
174
+ "grad_norm": 3.375,
175
+ "learning_rate": 4.800000000000001e-06,
176
+ "loss": 1.0236,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.41750167000668004,
181
+ "grad_norm": 1.6953125,
182
+ "learning_rate": 5e-06,
183
+ "loss": 1.0158,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.43420173680694724,
188
+ "grad_norm": 2.265625,
189
+ "learning_rate": 5.2e-06,
190
+ "loss": 1.0127,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.45090180360721444,
195
+ "grad_norm": 2.984375,
196
+ "learning_rate": 5.400000000000001e-06,
197
+ "loss": 1.0234,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.46760187040748163,
202
+ "grad_norm": 1.9609375,
203
+ "learning_rate": 5.600000000000001e-06,
204
+ "loss": 0.993,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.48430193720774883,
209
+ "grad_norm": 2.0625,
210
+ "learning_rate": 5.8e-06,
211
+ "loss": 1.0006,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.501002004008016,
216
+ "grad_norm": 1.0078125,
217
+ "learning_rate": 6e-06,
218
+ "loss": 1.0038,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 0.5177020708082832,
223
+ "grad_norm": 2.5625,
224
+ "learning_rate": 6.200000000000001e-06,
225
+ "loss": 0.9896,
226
+ "step": 1550
227
+ },
228
+ {
229
+ "epoch": 0.5344021376085505,
230
+ "grad_norm": 2.40625,
231
+ "learning_rate": 6.4000000000000006e-06,
232
+ "loss": 1.0005,
233
+ "step": 1600
234
+ },
235
+ {
236
+ "epoch": 0.5511022044088176,
237
+ "grad_norm": 1.546875,
238
+ "learning_rate": 6.600000000000001e-06,
239
+ "loss": 1.0176,
240
+ "step": 1650
241
+ },
242
+ {
243
+ "epoch": 0.5678022712090849,
244
+ "grad_norm": 1.8828125,
245
+ "learning_rate": 6.800000000000001e-06,
246
+ "loss": 0.9663,
247
+ "step": 1700
248
+ },
249
+ {
250
+ "epoch": 0.584502338009352,
251
+ "grad_norm": 1.1796875,
252
+ "learning_rate": 7e-06,
253
+ "loss": 0.9879,
254
+ "step": 1750
255
+ },
256
+ {
257
+ "epoch": 0.6012024048096193,
258
+ "grad_norm": 2.421875,
259
+ "learning_rate": 7.2000000000000005e-06,
260
+ "loss": 0.9724,
261
+ "step": 1800
262
+ },
263
+ {
264
+ "epoch": 0.6179024716098864,
265
+ "grad_norm": 1.5,
266
+ "learning_rate": 7.4e-06,
267
+ "loss": 0.9754,
268
+ "step": 1850
269
+ },
270
+ {
271
+ "epoch": 0.6346025384101537,
272
+ "grad_norm": 2.046875,
273
+ "learning_rate": 7.600000000000001e-06,
274
+ "loss": 0.9884,
275
+ "step": 1900
276
+ },
277
+ {
278
+ "epoch": 0.6513026052104208,
279
+ "grad_norm": 2.59375,
280
+ "learning_rate": 7.800000000000002e-06,
281
+ "loss": 0.9916,
282
+ "step": 1950
283
+ },
284
+ {
285
+ "epoch": 0.6680026720106881,
286
+ "grad_norm": 1.8984375,
287
+ "learning_rate": 8.000000000000001e-06,
288
+ "loss": 0.9821,
289
+ "step": 2000
290
+ },
291
+ {
292
+ "epoch": 0.6847027388109552,
293
+ "grad_norm": 1.484375,
294
+ "learning_rate": 8.2e-06,
295
+ "loss": 0.977,
296
+ "step": 2050
297
+ },
298
+ {
299
+ "epoch": 0.7014028056112225,
300
+ "grad_norm": 0.921875,
301
+ "learning_rate": 8.400000000000001e-06,
302
+ "loss": 0.9736,
303
+ "step": 2100
304
+ },
305
+ {
306
+ "epoch": 0.7181028724114896,
307
+ "grad_norm": 1.9609375,
308
+ "learning_rate": 8.6e-06,
309
+ "loss": 0.9545,
310
+ "step": 2150
311
+ },
312
+ {
313
+ "epoch": 0.7348029392117569,
314
+ "grad_norm": 2.125,
315
+ "learning_rate": 8.8e-06,
316
+ "loss": 0.9588,
317
+ "step": 2200
318
+ },
319
+ {
320
+ "epoch": 0.751503006012024,
321
+ "grad_norm": 1.8125,
322
+ "learning_rate": 9e-06,
323
+ "loss": 0.9655,
324
+ "step": 2250
325
+ },
326
+ {
327
+ "epoch": 0.7682030728122913,
328
+ "grad_norm": 1.25,
329
+ "learning_rate": 9.200000000000002e-06,
330
+ "loss": 0.9768,
331
+ "step": 2300
332
+ },
333
+ {
334
+ "epoch": 0.7849031396125584,
335
+ "grad_norm": 2.34375,
336
+ "learning_rate": 9.4e-06,
337
+ "loss": 0.9613,
338
+ "step": 2350
339
+ },
340
+ {
341
+ "epoch": 0.8016032064128257,
342
+ "grad_norm": 1.875,
343
+ "learning_rate": 9.600000000000001e-06,
344
+ "loss": 0.9656,
345
+ "step": 2400
346
+ },
347
+ {
348
+ "epoch": 0.8183032732130928,
349
+ "grad_norm": 1.3359375,
350
+ "learning_rate": 9.800000000000001e-06,
351
+ "loss": 0.9628,
352
+ "step": 2450
353
+ },
354
+ {
355
+ "epoch": 0.8350033400133601,
356
+ "grad_norm": 1.734375,
357
+ "learning_rate": 1e-05,
358
+ "loss": 0.9661,
359
+ "step": 2500
360
+ },
361
+ {
362
+ "epoch": 0.8517034068136272,
363
+ "grad_norm": 1.3046875,
364
+ "learning_rate": 1.02e-05,
365
+ "loss": 0.9763,
366
+ "step": 2550
367
+ },
368
+ {
369
+ "epoch": 0.8684034736138945,
370
+ "grad_norm": 1.3828125,
371
+ "learning_rate": 1.04e-05,
372
+ "loss": 0.9674,
373
+ "step": 2600
374
+ },
375
+ {
376
+ "epoch": 0.8851035404141616,
377
+ "grad_norm": 2.828125,
378
+ "learning_rate": 1.0600000000000002e-05,
379
+ "loss": 0.9804,
380
+ "step": 2650
381
+ },
382
+ {
383
+ "epoch": 0.9018036072144289,
384
+ "grad_norm": 3.53125,
385
+ "learning_rate": 1.0800000000000002e-05,
386
+ "loss": 0.9814,
387
+ "step": 2700
388
+ },
389
+ {
390
+ "epoch": 0.918503674014696,
391
+ "grad_norm": 1.5078125,
392
+ "learning_rate": 1.1000000000000001e-05,
393
+ "loss": 0.9553,
394
+ "step": 2750
395
+ },
396
+ {
397
+ "epoch": 0.9352037408149633,
398
+ "grad_norm": 1.796875,
399
+ "learning_rate": 1.1200000000000001e-05,
400
+ "loss": 0.9595,
401
+ "step": 2800
402
+ },
403
+ {
404
+ "epoch": 0.9519038076152304,
405
+ "grad_norm": 1.5078125,
406
+ "learning_rate": 1.14e-05,
407
+ "loss": 0.9364,
408
+ "step": 2850
409
+ },
410
+ {
411
+ "epoch": 0.9686038744154977,
412
+ "grad_norm": 3.28125,
413
+ "learning_rate": 1.16e-05,
414
+ "loss": 0.9597,
415
+ "step": 2900
416
+ },
417
+ {
418
+ "epoch": 0.9853039412157648,
419
+ "grad_norm": 1.609375,
420
+ "learning_rate": 1.18e-05,
421
+ "loss": 0.938,
422
+ "step": 2950
423
+ },
424
+ {
425
+ "epoch": 1.0,
426
+ "eval_loss": 0.945128321647644,
427
+ "eval_runtime": 100.8455,
428
+ "eval_samples_per_second": 15.985,
429
+ "eval_steps_per_second": 1.002,
430
+ "step": 2994
431
+ }
432
+ ],
433
+ "logging_steps": 50,
434
+ "max_steps": 5988,
435
+ "num_input_tokens_seen": 0,
436
+ "num_train_epochs": 2,
437
+ "save_steps": 500,
438
+ "stateful_callbacks": {
439
+ "TrainerControl": {
440
+ "args": {
441
+ "should_epoch_stop": false,
442
+ "should_evaluate": false,
443
+ "should_log": false,
444
+ "should_save": true,
445
+ "should_training_stop": false
446
+ },
447
+ "attributes": {}
448
+ }
449
+ },
450
+ "total_flos": 3.8655046824493056e+17,
451
+ "train_batch_size": 16,
452
+ "trial_name": null,
453
+ "trial_params": null
454
+ }
checkpoint-2994/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fee32f064590f4dbedd46c2ae13a662c26baa4367c87d2089d6e6d34cdcae7b
3
+ size 5432
checkpoint-5988/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/phi-1_5",
3
+ "architectures": [
4
+ "PhiForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": null,
8
+ "embd_pdrop": 0.0,
9
+ "eos_token_id": null,
10
+ "hidden_act": "gelu_new",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8192,
14
+ "layer_norm_eps": 1e-05,
15
+ "max_position_embeddings": 2048,
16
+ "model_type": "phi",
17
+ "num_attention_heads": 32,
18
+ "num_hidden_layers": 24,
19
+ "num_key_value_heads": 32,
20
+ "partial_rotary_factor": 0.5,
21
+ "qk_layernorm": false,
22
+ "resid_pdrop": 0.0,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000.0,
25
+ "tie_word_embeddings": false,
26
+ "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.45.0",
28
+ "use_cache": false,
29
+ "vocab_size": 51200
30
+ }
checkpoint-5988/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.45.0"
4
+ }
checkpoint-5988/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:659a8db5e3e06113bc4e343fc87ae3b103823d2ff6efbe0c6eddafefd06183b9
3
+ size 2836579040
checkpoint-5988/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5540fe647696a7c1294b5f87f8b2b9167bca95a5e8bd61272cf1c691bc1d4285
3
+ size 11346377382
checkpoint-5988/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d138cfe3a4adf21f048848ee35837c9a757a0a3616ff7adbb45b69aac247435
3
+ size 14244
checkpoint-5988/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9e7aec8d0b74157d11a4dcca01d36a7ec3df8af858b2081233b73e3c09150cc
3
+ size 1064
checkpoint-5988/trainer_state.json ADDED
@@ -0,0 +1,882 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.877801239490509,
3
+ "best_model_checkpoint": "./checkpoints/ultrafeedback_binarized/phi-1_5-ultrafeedback_binarized-lambda0.25-ORPO-18-16-19/checkpoint-5988",
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 5988,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.016700066800267203,
13
+ "grad_norm": 5.34375,
14
+ "learning_rate": 2.0000000000000002e-07,
15
+ "loss": 1.2526,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.033400133600534405,
20
+ "grad_norm": 3.453125,
21
+ "learning_rate": 4.0000000000000003e-07,
22
+ "loss": 1.2115,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.050100200400801605,
27
+ "grad_norm": 2.953125,
28
+ "learning_rate": 6.000000000000001e-07,
29
+ "loss": 1.2056,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.06680026720106881,
34
+ "grad_norm": 2.3125,
35
+ "learning_rate": 8.000000000000001e-07,
36
+ "loss": 1.2425,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.08350033400133601,
41
+ "grad_norm": 1.9921875,
42
+ "learning_rate": 1.0000000000000002e-06,
43
+ "loss": 1.2355,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.10020040080160321,
48
+ "grad_norm": 2.484375,
49
+ "learning_rate": 1.2000000000000002e-06,
50
+ "loss": 1.2408,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.11690046760187041,
55
+ "grad_norm": 2.453125,
56
+ "learning_rate": 1.4000000000000001e-06,
57
+ "loss": 1.2136,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.13360053440213762,
62
+ "grad_norm": 2.328125,
63
+ "learning_rate": 1.6000000000000001e-06,
64
+ "loss": 1.2241,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.15030060120240482,
69
+ "grad_norm": 3.859375,
70
+ "learning_rate": 1.8000000000000001e-06,
71
+ "loss": 1.213,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.16700066800267202,
76
+ "grad_norm": 2.90625,
77
+ "learning_rate": 2.0000000000000003e-06,
78
+ "loss": 1.2212,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.18370073480293922,
83
+ "grad_norm": 1.828125,
84
+ "learning_rate": 2.2e-06,
85
+ "loss": 1.1722,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.20040080160320642,
90
+ "grad_norm": 2.953125,
91
+ "learning_rate": 2.4000000000000003e-06,
92
+ "loss": 1.1541,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.21710086840347362,
97
+ "grad_norm": 4.3125,
98
+ "learning_rate": 2.6e-06,
99
+ "loss": 1.1767,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.23380093520374082,
104
+ "grad_norm": 2.828125,
105
+ "learning_rate": 2.8000000000000003e-06,
106
+ "loss": 1.0593,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.250501002004008,
111
+ "grad_norm": 1.7421875,
112
+ "learning_rate": 3e-06,
113
+ "loss": 1.128,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.26720106880427524,
118
+ "grad_norm": 2.34375,
119
+ "learning_rate": 3.2000000000000003e-06,
120
+ "loss": 1.1307,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.28390113560454244,
125
+ "grad_norm": 2.390625,
126
+ "learning_rate": 3.4000000000000005e-06,
127
+ "loss": 1.0742,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.30060120240480964,
132
+ "grad_norm": 1.921875,
133
+ "learning_rate": 3.6000000000000003e-06,
134
+ "loss": 1.078,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.31730126920507684,
139
+ "grad_norm": 2.4375,
140
+ "learning_rate": 3.8000000000000005e-06,
141
+ "loss": 1.063,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.33400133600534404,
146
+ "grad_norm": 2.59375,
147
+ "learning_rate": 4.000000000000001e-06,
148
+ "loss": 1.0211,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.35070140280561124,
153
+ "grad_norm": 1.7890625,
154
+ "learning_rate": 4.2000000000000004e-06,
155
+ "loss": 1.0235,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.36740146960587844,
160
+ "grad_norm": 2.46875,
161
+ "learning_rate": 4.4e-06,
162
+ "loss": 1.0346,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.38410153640614564,
167
+ "grad_norm": 1.0625,
168
+ "learning_rate": 4.600000000000001e-06,
169
+ "loss": 1.038,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.40080160320641284,
174
+ "grad_norm": 3.375,
175
+ "learning_rate": 4.800000000000001e-06,
176
+ "loss": 1.0236,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.41750167000668004,
181
+ "grad_norm": 1.6953125,
182
+ "learning_rate": 5e-06,
183
+ "loss": 1.0158,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.43420173680694724,
188
+ "grad_norm": 2.265625,
189
+ "learning_rate": 5.2e-06,
190
+ "loss": 1.0127,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.45090180360721444,
195
+ "grad_norm": 2.984375,
196
+ "learning_rate": 5.400000000000001e-06,
197
+ "loss": 1.0234,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.46760187040748163,
202
+ "grad_norm": 1.9609375,
203
+ "learning_rate": 5.600000000000001e-06,
204
+ "loss": 0.993,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.48430193720774883,
209
+ "grad_norm": 2.0625,
210
+ "learning_rate": 5.8e-06,
211
+ "loss": 1.0006,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.501002004008016,
216
+ "grad_norm": 1.0078125,
217
+ "learning_rate": 6e-06,
218
+ "loss": 1.0038,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 0.5177020708082832,
223
+ "grad_norm": 2.5625,
224
+ "learning_rate": 6.200000000000001e-06,
225
+ "loss": 0.9896,
226
+ "step": 1550
227
+ },
228
+ {
229
+ "epoch": 0.5344021376085505,
230
+ "grad_norm": 2.40625,
231
+ "learning_rate": 6.4000000000000006e-06,
232
+ "loss": 1.0005,
233
+ "step": 1600
234
+ },
235
+ {
236
+ "epoch": 0.5511022044088176,
237
+ "grad_norm": 1.546875,
238
+ "learning_rate": 6.600000000000001e-06,
239
+ "loss": 1.0176,
240
+ "step": 1650
241
+ },
242
+ {
243
+ "epoch": 0.5678022712090849,
244
+ "grad_norm": 1.8828125,
245
+ "learning_rate": 6.800000000000001e-06,
246
+ "loss": 0.9663,
247
+ "step": 1700
248
+ },
249
+ {
250
+ "epoch": 0.584502338009352,
251
+ "grad_norm": 1.1796875,
252
+ "learning_rate": 7e-06,
253
+ "loss": 0.9879,
254
+ "step": 1750
255
+ },
256
+ {
257
+ "epoch": 0.6012024048096193,
258
+ "grad_norm": 2.421875,
259
+ "learning_rate": 7.2000000000000005e-06,
260
+ "loss": 0.9724,
261
+ "step": 1800
262
+ },
263
+ {
264
+ "epoch": 0.6179024716098864,
265
+ "grad_norm": 1.5,
266
+ "learning_rate": 7.4e-06,
267
+ "loss": 0.9754,
268
+ "step": 1850
269
+ },
270
+ {
271
+ "epoch": 0.6346025384101537,
272
+ "grad_norm": 2.046875,
273
+ "learning_rate": 7.600000000000001e-06,
274
+ "loss": 0.9884,
275
+ "step": 1900
276
+ },
277
+ {
278
+ "epoch": 0.6513026052104208,
279
+ "grad_norm": 2.59375,
280
+ "learning_rate": 7.800000000000002e-06,
281
+ "loss": 0.9916,
282
+ "step": 1950
283
+ },
284
+ {
285
+ "epoch": 0.6680026720106881,
286
+ "grad_norm": 1.8984375,
287
+ "learning_rate": 8.000000000000001e-06,
288
+ "loss": 0.9821,
289
+ "step": 2000
290
+ },
291
+ {
292
+ "epoch": 0.6847027388109552,
293
+ "grad_norm": 1.484375,
294
+ "learning_rate": 8.2e-06,
295
+ "loss": 0.977,
296
+ "step": 2050
297
+ },
298
+ {
299
+ "epoch": 0.7014028056112225,
300
+ "grad_norm": 0.921875,
301
+ "learning_rate": 8.400000000000001e-06,
302
+ "loss": 0.9736,
303
+ "step": 2100
304
+ },
305
+ {
306
+ "epoch": 0.7181028724114896,
307
+ "grad_norm": 1.9609375,
308
+ "learning_rate": 8.6e-06,
309
+ "loss": 0.9545,
310
+ "step": 2150
311
+ },
312
+ {
313
+ "epoch": 0.7348029392117569,
314
+ "grad_norm": 2.125,
315
+ "learning_rate": 8.8e-06,
316
+ "loss": 0.9588,
317
+ "step": 2200
318
+ },
319
+ {
320
+ "epoch": 0.751503006012024,
321
+ "grad_norm": 1.8125,
322
+ "learning_rate": 9e-06,
323
+ "loss": 0.9655,
324
+ "step": 2250
325
+ },
326
+ {
327
+ "epoch": 0.7682030728122913,
328
+ "grad_norm": 1.25,
329
+ "learning_rate": 9.200000000000002e-06,
330
+ "loss": 0.9768,
331
+ "step": 2300
332
+ },
333
+ {
334
+ "epoch": 0.7849031396125584,
335
+ "grad_norm": 2.34375,
336
+ "learning_rate": 9.4e-06,
337
+ "loss": 0.9613,
338
+ "step": 2350
339
+ },
340
+ {
341
+ "epoch": 0.8016032064128257,
342
+ "grad_norm": 1.875,
343
+ "learning_rate": 9.600000000000001e-06,
344
+ "loss": 0.9656,
345
+ "step": 2400
346
+ },
347
+ {
348
+ "epoch": 0.8183032732130928,
349
+ "grad_norm": 1.3359375,
350
+ "learning_rate": 9.800000000000001e-06,
351
+ "loss": 0.9628,
352
+ "step": 2450
353
+ },
354
+ {
355
+ "epoch": 0.8350033400133601,
356
+ "grad_norm": 1.734375,
357
+ "learning_rate": 1e-05,
358
+ "loss": 0.9661,
359
+ "step": 2500
360
+ },
361
+ {
362
+ "epoch": 0.8517034068136272,
363
+ "grad_norm": 1.3046875,
364
+ "learning_rate": 1.02e-05,
365
+ "loss": 0.9763,
366
+ "step": 2550
367
+ },
368
+ {
369
+ "epoch": 0.8684034736138945,
370
+ "grad_norm": 1.3828125,
371
+ "learning_rate": 1.04e-05,
372
+ "loss": 0.9674,
373
+ "step": 2600
374
+ },
375
+ {
376
+ "epoch": 0.8851035404141616,
377
+ "grad_norm": 2.828125,
378
+ "learning_rate": 1.0600000000000002e-05,
379
+ "loss": 0.9804,
380
+ "step": 2650
381
+ },
382
+ {
383
+ "epoch": 0.9018036072144289,
384
+ "grad_norm": 3.53125,
385
+ "learning_rate": 1.0800000000000002e-05,
386
+ "loss": 0.9814,
387
+ "step": 2700
388
+ },
389
+ {
390
+ "epoch": 0.918503674014696,
391
+ "grad_norm": 1.5078125,
392
+ "learning_rate": 1.1000000000000001e-05,
393
+ "loss": 0.9553,
394
+ "step": 2750
395
+ },
396
+ {
397
+ "epoch": 0.9352037408149633,
398
+ "grad_norm": 1.796875,
399
+ "learning_rate": 1.1200000000000001e-05,
400
+ "loss": 0.9595,
401
+ "step": 2800
402
+ },
403
+ {
404
+ "epoch": 0.9519038076152304,
405
+ "grad_norm": 1.5078125,
406
+ "learning_rate": 1.14e-05,
407
+ "loss": 0.9364,
408
+ "step": 2850
409
+ },
410
+ {
411
+ "epoch": 0.9686038744154977,
412
+ "grad_norm": 3.28125,
413
+ "learning_rate": 1.16e-05,
414
+ "loss": 0.9597,
415
+ "step": 2900
416
+ },
417
+ {
418
+ "epoch": 0.9853039412157648,
419
+ "grad_norm": 1.609375,
420
+ "learning_rate": 1.18e-05,
421
+ "loss": 0.938,
422
+ "step": 2950
423
+ },
424
+ {
425
+ "epoch": 1.0,
426
+ "eval_loss": 0.945128321647644,
427
+ "eval_runtime": 100.8455,
428
+ "eval_samples_per_second": 15.985,
429
+ "eval_steps_per_second": 1.002,
430
+ "step": 2994
431
+ },
432
+ {
433
+ "epoch": 1.002004008016032,
434
+ "grad_norm": 2.703125,
435
+ "learning_rate": 1.2e-05,
436
+ "loss": 0.9488,
437
+ "step": 3000
438
+ },
439
+ {
440
+ "epoch": 1.0187040748162992,
441
+ "grad_norm": 6.40625,
442
+ "learning_rate": 1.22e-05,
443
+ "loss": 0.9558,
444
+ "step": 3050
445
+ },
446
+ {
447
+ "epoch": 1.0354041416165665,
448
+ "grad_norm": 1.2890625,
449
+ "learning_rate": 1.2400000000000002e-05,
450
+ "loss": 0.9402,
451
+ "step": 3100
452
+ },
453
+ {
454
+ "epoch": 1.0521042084168337,
455
+ "grad_norm": 2.625,
456
+ "learning_rate": 1.2600000000000001e-05,
457
+ "loss": 0.9402,
458
+ "step": 3150
459
+ },
460
+ {
461
+ "epoch": 1.0688042752171008,
462
+ "grad_norm": 2.546875,
463
+ "learning_rate": 1.2800000000000001e-05,
464
+ "loss": 0.9116,
465
+ "step": 3200
466
+ },
467
+ {
468
+ "epoch": 1.085504342017368,
469
+ "grad_norm": 3.84375,
470
+ "learning_rate": 1.3000000000000001e-05,
471
+ "loss": 0.9298,
472
+ "step": 3250
473
+ },
474
+ {
475
+ "epoch": 1.1022044088176353,
476
+ "grad_norm": 1.921875,
477
+ "learning_rate": 1.3200000000000002e-05,
478
+ "loss": 0.9347,
479
+ "step": 3300
480
+ },
481
+ {
482
+ "epoch": 1.1189044756179025,
483
+ "grad_norm": 2.453125,
484
+ "learning_rate": 1.3400000000000002e-05,
485
+ "loss": 0.9245,
486
+ "step": 3350
487
+ },
488
+ {
489
+ "epoch": 1.1356045424181698,
490
+ "grad_norm": 2.796875,
491
+ "learning_rate": 1.3600000000000002e-05,
492
+ "loss": 0.962,
493
+ "step": 3400
494
+ },
495
+ {
496
+ "epoch": 1.1523046092184368,
497
+ "grad_norm": 2.109375,
498
+ "learning_rate": 1.38e-05,
499
+ "loss": 0.9058,
500
+ "step": 3450
501
+ },
502
+ {
503
+ "epoch": 1.169004676018704,
504
+ "grad_norm": 2.53125,
505
+ "learning_rate": 1.4e-05,
506
+ "loss": 0.9173,
507
+ "step": 3500
508
+ },
509
+ {
510
+ "epoch": 1.1857047428189713,
511
+ "grad_norm": 3.34375,
512
+ "learning_rate": 1.4200000000000001e-05,
513
+ "loss": 0.9079,
514
+ "step": 3550
515
+ },
516
+ {
517
+ "epoch": 1.2024048096192386,
518
+ "grad_norm": 2.515625,
519
+ "learning_rate": 1.4400000000000001e-05,
520
+ "loss": 0.9239,
521
+ "step": 3600
522
+ },
523
+ {
524
+ "epoch": 1.2191048764195056,
525
+ "grad_norm": 1.6015625,
526
+ "learning_rate": 1.46e-05,
527
+ "loss": 0.9147,
528
+ "step": 3650
529
+ },
530
+ {
531
+ "epoch": 1.2358049432197729,
532
+ "grad_norm": 2.59375,
533
+ "learning_rate": 1.48e-05,
534
+ "loss": 0.9367,
535
+ "step": 3700
536
+ },
537
+ {
538
+ "epoch": 1.25250501002004,
539
+ "grad_norm": 2.890625,
540
+ "learning_rate": 1.5000000000000002e-05,
541
+ "loss": 0.9116,
542
+ "step": 3750
543
+ },
544
+ {
545
+ "epoch": 1.2692050768203074,
546
+ "grad_norm": 2.765625,
547
+ "learning_rate": 1.5200000000000002e-05,
548
+ "loss": 0.9198,
549
+ "step": 3800
550
+ },
551
+ {
552
+ "epoch": 1.2859051436205746,
553
+ "grad_norm": 3.046875,
554
+ "learning_rate": 1.54e-05,
555
+ "loss": 0.908,
556
+ "step": 3850
557
+ },
558
+ {
559
+ "epoch": 1.3026052104208417,
560
+ "grad_norm": 1.953125,
561
+ "learning_rate": 1.5600000000000003e-05,
562
+ "loss": 0.908,
563
+ "step": 3900
564
+ },
565
+ {
566
+ "epoch": 1.319305277221109,
567
+ "grad_norm": 1.578125,
568
+ "learning_rate": 1.58e-05,
569
+ "loss": 0.9109,
570
+ "step": 3950
571
+ },
572
+ {
573
+ "epoch": 1.3360053440213762,
574
+ "grad_norm": 2.421875,
575
+ "learning_rate": 1.6000000000000003e-05,
576
+ "loss": 0.9148,
577
+ "step": 4000
578
+ },
579
+ {
580
+ "epoch": 1.3527054108216432,
581
+ "grad_norm": 4.28125,
582
+ "learning_rate": 1.62e-05,
583
+ "loss": 0.8995,
584
+ "step": 4050
585
+ },
586
+ {
587
+ "epoch": 1.3694054776219104,
588
+ "grad_norm": 2.765625,
589
+ "learning_rate": 1.64e-05,
590
+ "loss": 0.9177,
591
+ "step": 4100
592
+ },
593
+ {
594
+ "epoch": 1.3861055444221777,
595
+ "grad_norm": 2.921875,
596
+ "learning_rate": 1.66e-05,
597
+ "loss": 0.9138,
598
+ "step": 4150
599
+ },
600
+ {
601
+ "epoch": 1.402805611222445,
602
+ "grad_norm": 2.46875,
603
+ "learning_rate": 1.6800000000000002e-05,
604
+ "loss": 0.9378,
605
+ "step": 4200
606
+ },
607
+ {
608
+ "epoch": 1.4195056780227122,
609
+ "grad_norm": 2.578125,
610
+ "learning_rate": 1.7e-05,
611
+ "loss": 0.8952,
612
+ "step": 4250
613
+ },
614
+ {
615
+ "epoch": 1.4362057448229792,
616
+ "grad_norm": 2.890625,
617
+ "learning_rate": 1.72e-05,
618
+ "loss": 0.9045,
619
+ "step": 4300
620
+ },
621
+ {
622
+ "epoch": 1.4529058116232465,
623
+ "grad_norm": 2.375,
624
+ "learning_rate": 1.7400000000000003e-05,
625
+ "loss": 0.942,
626
+ "step": 4350
627
+ },
628
+ {
629
+ "epoch": 1.4696058784235138,
630
+ "grad_norm": 2.296875,
631
+ "learning_rate": 1.76e-05,
632
+ "loss": 0.9018,
633
+ "step": 4400
634
+ },
635
+ {
636
+ "epoch": 1.4863059452237808,
637
+ "grad_norm": 2.546875,
638
+ "learning_rate": 1.7800000000000002e-05,
639
+ "loss": 0.886,
640
+ "step": 4450
641
+ },
642
+ {
643
+ "epoch": 1.503006012024048,
644
+ "grad_norm": 2.40625,
645
+ "learning_rate": 1.8e-05,
646
+ "loss": 0.9152,
647
+ "step": 4500
648
+ },
649
+ {
650
+ "epoch": 1.5197060788243153,
651
+ "grad_norm": 1.375,
652
+ "learning_rate": 1.8200000000000002e-05,
653
+ "loss": 0.8884,
654
+ "step": 4550
655
+ },
656
+ {
657
+ "epoch": 1.5364061456245826,
658
+ "grad_norm": 2.703125,
659
+ "learning_rate": 1.8400000000000003e-05,
660
+ "loss": 0.8905,
661
+ "step": 4600
662
+ },
663
+ {
664
+ "epoch": 1.5531062124248498,
665
+ "grad_norm": 3.15625,
666
+ "learning_rate": 1.86e-05,
667
+ "loss": 0.9163,
668
+ "step": 4650
669
+ },
670
+ {
671
+ "epoch": 1.569806279225117,
672
+ "grad_norm": 2.921875,
673
+ "learning_rate": 1.88e-05,
674
+ "loss": 0.9196,
675
+ "step": 4700
676
+ },
677
+ {
678
+ "epoch": 1.586506346025384,
679
+ "grad_norm": 2.296875,
680
+ "learning_rate": 1.9e-05,
681
+ "loss": 0.88,
682
+ "step": 4750
683
+ },
684
+ {
685
+ "epoch": 1.6032064128256514,
686
+ "grad_norm": 2.671875,
687
+ "learning_rate": 1.9200000000000003e-05,
688
+ "loss": 0.887,
689
+ "step": 4800
690
+ },
691
+ {
692
+ "epoch": 1.6199064796259184,
693
+ "grad_norm": 2.046875,
694
+ "learning_rate": 1.94e-05,
695
+ "loss": 0.8727,
696
+ "step": 4850
697
+ },
698
+ {
699
+ "epoch": 1.6366065464261856,
700
+ "grad_norm": 1.953125,
701
+ "learning_rate": 1.9600000000000002e-05,
702
+ "loss": 0.8891,
703
+ "step": 4900
704
+ },
705
+ {
706
+ "epoch": 1.653306613226453,
707
+ "grad_norm": 2.890625,
708
+ "learning_rate": 1.98e-05,
709
+ "loss": 0.9095,
710
+ "step": 4950
711
+ },
712
+ {
713
+ "epoch": 1.6700066800267201,
714
+ "grad_norm": 2.28125,
715
+ "learning_rate": 2e-05,
716
+ "loss": 0.8675,
717
+ "step": 5000
718
+ },
719
+ {
720
+ "epoch": 1.6867067468269874,
721
+ "grad_norm": 4.0,
722
+ "learning_rate": 1.9873880897766597e-05,
723
+ "loss": 0.8988,
724
+ "step": 5050
725
+ },
726
+ {
727
+ "epoch": 1.7034068136272547,
728
+ "grad_norm": 1.5859375,
729
+ "learning_rate": 1.949870479665602e-05,
730
+ "loss": 0.8884,
731
+ "step": 5100
732
+ },
733
+ {
734
+ "epoch": 1.7201068804275217,
735
+ "grad_norm": 2.5,
736
+ "learning_rate": 1.888393507127856e-05,
737
+ "loss": 0.8995,
738
+ "step": 5150
739
+ },
740
+ {
741
+ "epoch": 1.736806947227789,
742
+ "grad_norm": 4.125,
743
+ "learning_rate": 1.8045078562803203e-05,
744
+ "loss": 0.8793,
745
+ "step": 5200
746
+ },
747
+ {
748
+ "epoch": 1.753507014028056,
749
+ "grad_norm": 1.3046875,
750
+ "learning_rate": 1.7003294437180254e-05,
751
+ "loss": 0.889,
752
+ "step": 5250
753
+ },
754
+ {
755
+ "epoch": 1.7702070808283232,
756
+ "grad_norm": 1.796875,
757
+ "learning_rate": 1.5784860470138633e-05,
758
+ "loss": 0.8857,
759
+ "step": 5300
760
+ },
761
+ {
762
+ "epoch": 1.7869071476285905,
763
+ "grad_norm": 2.5,
764
+ "learning_rate": 1.4420510221289137e-05,
765
+ "loss": 0.8643,
766
+ "step": 5350
767
+ },
768
+ {
769
+ "epoch": 1.8036072144288577,
770
+ "grad_norm": 1.6953125,
771
+ "learning_rate": 1.2944657816335124e-05,
772
+ "loss": 0.8841,
773
+ "step": 5400
774
+ },
775
+ {
776
+ "epoch": 1.820307281229125,
777
+ "grad_norm": 1.640625,
778
+ "learning_rate": 1.139452989134496e-05,
779
+ "loss": 0.8548,
780
+ "step": 5450
781
+ },
782
+ {
783
+ "epoch": 1.8370073480293923,
784
+ "grad_norm": 1.6328125,
785
+ "learning_rate": 9.809226594767979e-06,
786
+ "loss": 0.8762,
787
+ "step": 5500
788
+ },
789
+ {
790
+ "epoch": 1.8537074148296593,
791
+ "grad_norm": 2.0,
792
+ "learning_rate": 8.228735332310575e-06,
793
+ "loss": 0.8726,
794
+ "step": 5550
795
+ },
796
+ {
797
+ "epoch": 1.8704074816299265,
798
+ "grad_norm": 2.28125,
799
+ "learning_rate": 6.692922131794517e-06,
800
+ "loss": 0.8995,
801
+ "step": 5600
802
+ },
803
+ {
804
+ "epoch": 1.8871075484301936,
805
+ "grad_norm": 3.09375,
806
+ "learning_rate": 5.240526069629265e-06,
807
+ "loss": 0.8944,
808
+ "step": 5650
809
+ },
810
+ {
811
+ "epoch": 1.9038076152304608,
812
+ "grad_norm": 1.8203125,
813
+ "learning_rate": 3.908182123304344e-06,
814
+ "loss": 0.8777,
815
+ "step": 5700
816
+ },
817
+ {
818
+ "epoch": 1.920507682030728,
819
+ "grad_norm": 1.2421875,
820
+ "learning_rate": 2.729497097295075e-06,
821
+ "loss": 0.8733,
822
+ "step": 5750
823
+ },
824
+ {
825
+ "epoch": 1.9372077488309953,
826
+ "grad_norm": 2.71875,
827
+ "learning_rate": 1.7342019310607062e-06,
828
+ "loss": 0.8705,
829
+ "step": 5800
830
+ },
831
+ {
832
+ "epoch": 1.9539078156312626,
833
+ "grad_norm": 1.9375,
834
+ "learning_rate": 9.474017711657835e-07,
835
+ "loss": 0.8602,
836
+ "step": 5850
837
+ },
838
+ {
839
+ "epoch": 1.9706078824315298,
840
+ "grad_norm": 2.765625,
841
+ "learning_rate": 3.889427235709153e-07,
842
+ "loss": 0.8755,
843
+ "step": 5900
844
+ },
845
+ {
846
+ "epoch": 1.9873079492317969,
847
+ "grad_norm": 2.515625,
848
+ "learning_rate": 7.291125901946027e-08,
849
+ "loss": 0.8718,
850
+ "step": 5950
851
+ },
852
+ {
853
+ "epoch": 2.0,
854
+ "eval_loss": 0.877801239490509,
855
+ "eval_runtime": 101.041,
856
+ "eval_samples_per_second": 15.954,
857
+ "eval_steps_per_second": 1.0,
858
+ "step": 5988
859
+ }
860
+ ],
861
+ "logging_steps": 50,
862
+ "max_steps": 5988,
863
+ "num_input_tokens_seen": 0,
864
+ "num_train_epochs": 2,
865
+ "save_steps": 500,
866
+ "stateful_callbacks": {
867
+ "TrainerControl": {
868
+ "args": {
869
+ "should_epoch_stop": false,
870
+ "should_evaluate": false,
871
+ "should_log": false,
872
+ "should_save": true,
873
+ "should_training_stop": true
874
+ },
875
+ "attributes": {}
876
+ }
877
+ },
878
+ "total_flos": 7.731009364898611e+17,
879
+ "train_batch_size": 16,
880
+ "trial_name": null,
881
+ "trial_params": null
882
+ }
checkpoint-5988/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fee32f064590f4dbedd46c2ae13a662c26baa4367c87d2089d6e6d34cdcae7b
3
+ size 5432
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/phi-1_5",
3
+ "architectures": [
4
+ "PhiForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": null,
8
+ "embd_pdrop": 0.0,
9
+ "eos_token_id": null,
10
+ "hidden_act": "gelu_new",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8192,
14
+ "layer_norm_eps": 1e-05,
15
+ "max_position_embeddings": 2048,
16
+ "model_type": "phi",
17
+ "num_attention_heads": 32,
18
+ "num_hidden_layers": 24,
19
+ "num_key_value_heads": 32,
20
+ "partial_rotary_factor": 0.5,
21
+ "qk_layernorm": false,
22
+ "resid_pdrop": 0.0,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000.0,
25
+ "tie_word_embeddings": false,
26
+ "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.45.0",
28
+ "use_cache": false,
29
+ "vocab_size": 51200
30
+ }
generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.45.0"
4
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:659a8db5e3e06113bc4e343fc87ae3b103823d2ff6efbe0c6eddafefd06183b9
3
+ size 2836579040
special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "unk_token": "<|endoftext|>"
5
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<|endoftext|>",
4
+ "clean_up_tokenization_spaces": true,
5
+ "eos_token": "<|endoftext|>",
6
+ "model_max_length": 2048,
7
+ "tokenizer_class": "CodeGenTokenizer",
8
+ "unk_token": "<|endoftext|>"
9
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fee32f064590f4dbedd46c2ae13a662c26baa4367c87d2089d6e6d34cdcae7b
3
+ size 5432