Grogros commited on
Commit
571fa93
·
verified ·
1 Parent(s): 858a349

Training in progress, step 1500, checkpoint

Browse files
checkpoint-1500/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "meta-llama/Llama-3.2-1B-Instruct",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 128000,
9
+ "eos_token_id": [
10
+ 128001,
11
+ 128008,
12
+ 128009
13
+ ],
14
+ "head_dim": 64,
15
+ "hidden_act": "silu",
16
+ "hidden_size": 2048,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 8192,
19
+ "max_position_embeddings": 131072,
20
+ "mlp_bias": false,
21
+ "model_type": "llama",
22
+ "num_attention_heads": 32,
23
+ "num_hidden_layers": 16,
24
+ "num_key_value_heads": 8,
25
+ "pretraining_tp": 1,
26
+ "rms_norm_eps": 1e-05,
27
+ "rope_scaling": {
28
+ "factor": 32.0,
29
+ "high_freq_factor": 4.0,
30
+ "low_freq_factor": 1.0,
31
+ "original_max_position_embeddings": 8192,
32
+ "rope_type": "llama3"
33
+ },
34
+ "rope_theta": 500000.0,
35
+ "tie_word_embeddings": true,
36
+ "torch_dtype": "bfloat16",
37
+ "transformers_version": "4.46.3",
38
+ "use_cache": true,
39
+ "vocab_size": 128256
40
+ }
checkpoint-1500/generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 128000,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 128001,
6
+ 128008,
7
+ 128009
8
+ ],
9
+ "temperature": 0.6,
10
+ "top_p": 0.9,
11
+ "transformers_version": "4.46.3"
12
+ }
checkpoint-1500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06bb2c4d5d91910dc814e641e48609741e7009865298f28595da2d30edd96016
3
+ size 2471645608
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:170cb434a03ed2fd43f282a8b8979745fad01db35f937af429e456eb745ff3e5
3
+ size 3724602
checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5450255e6a9ab1eb65c5f2d4ad14e937c363f0fe9b8fbbad54746263bc4a8f15
3
+ size 14244
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b62a6982caa9b47207f64a7858dbd39e1fbcf0807c59221faa61c7a640f067d6
3
+ size 1064
checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,1083 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.09325,
5
+ "eval_steps": 500,
6
+ "global_step": 1500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.005,
13
+ "grad_norm": 402.0,
14
+ "learning_rate": 1.0000000000000002e-06,
15
+ "loss": 4.0977,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.01,
20
+ "grad_norm": 51.0,
21
+ "learning_rate": 2.0000000000000003e-06,
22
+ "loss": 4.0891,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.015,
27
+ "grad_norm": 50.0,
28
+ "learning_rate": 3e-06,
29
+ "loss": 3.4992,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.02,
34
+ "grad_norm": 23.5,
35
+ "learning_rate": 4.000000000000001e-06,
36
+ "loss": 2.9609,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.025,
41
+ "grad_norm": 16.5,
42
+ "learning_rate": 5e-06,
43
+ "loss": 3.1281,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.03,
48
+ "grad_norm": 20.375,
49
+ "learning_rate": 6e-06,
50
+ "loss": 3.0043,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.035,
55
+ "grad_norm": 18.125,
56
+ "learning_rate": 7e-06,
57
+ "loss": 2.8566,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.04,
62
+ "grad_norm": 42.0,
63
+ "learning_rate": 8.000000000000001e-06,
64
+ "loss": 3.1531,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.045,
69
+ "grad_norm": 17.25,
70
+ "learning_rate": 9e-06,
71
+ "loss": 2.6734,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.05,
76
+ "grad_norm": 13.6875,
77
+ "learning_rate": 1e-05,
78
+ "loss": 2.6406,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.055,
83
+ "grad_norm": 11.875,
84
+ "learning_rate": 1.1000000000000001e-05,
85
+ "loss": 2.7305,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.06,
90
+ "grad_norm": 15.5,
91
+ "learning_rate": 1.2e-05,
92
+ "loss": 2.6008,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.065,
97
+ "grad_norm": 12.9375,
98
+ "learning_rate": 1.3000000000000001e-05,
99
+ "loss": 2.5461,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.07,
104
+ "grad_norm": 14.4375,
105
+ "learning_rate": 1.4e-05,
106
+ "loss": 2.6531,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.075,
111
+ "grad_norm": 12.125,
112
+ "learning_rate": 1.5000000000000002e-05,
113
+ "loss": 2.3348,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.08,
118
+ "grad_norm": 13.25,
119
+ "learning_rate": 1.6000000000000003e-05,
120
+ "loss": 2.268,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.085,
125
+ "grad_norm": 13.4375,
126
+ "learning_rate": 1.7e-05,
127
+ "loss": 2.3805,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.09,
132
+ "grad_norm": 18.125,
133
+ "learning_rate": 1.8e-05,
134
+ "loss": 2.2859,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.095,
139
+ "grad_norm": 14.75,
140
+ "learning_rate": 1.9e-05,
141
+ "loss": 2.2238,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.1,
146
+ "grad_norm": 11.6875,
147
+ "learning_rate": 2e-05,
148
+ "loss": 2.2816,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.105,
153
+ "grad_norm": 18.875,
154
+ "learning_rate": 1.9998476951563914e-05,
155
+ "loss": 1.8926,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.11,
160
+ "grad_norm": 10.8125,
161
+ "learning_rate": 1.999390827019096e-05,
162
+ "loss": 2.0168,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.115,
167
+ "grad_norm": 30.0,
168
+ "learning_rate": 1.9986295347545738e-05,
169
+ "loss": 1.9996,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.12,
174
+ "grad_norm": 14.6875,
175
+ "learning_rate": 1.9975640502598243e-05,
176
+ "loss": 1.8109,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.125,
181
+ "grad_norm": 14.0625,
182
+ "learning_rate": 1.9961946980917457e-05,
183
+ "loss": 1.7777,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.13,
188
+ "grad_norm": 22.125,
189
+ "learning_rate": 1.9945218953682736e-05,
190
+ "loss": 1.7828,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.135,
195
+ "grad_norm": 28.75,
196
+ "learning_rate": 1.9925461516413224e-05,
197
+ "loss": 1.5926,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.14,
202
+ "grad_norm": 17.375,
203
+ "learning_rate": 1.9902680687415704e-05,
204
+ "loss": 1.7307,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.145,
209
+ "grad_norm": 14.4375,
210
+ "learning_rate": 1.9876883405951378e-05,
211
+ "loss": 1.6797,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.15,
216
+ "grad_norm": 45.0,
217
+ "learning_rate": 1.9848077530122083e-05,
218
+ "loss": 1.4248,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.155,
223
+ "grad_norm": 12.9375,
224
+ "learning_rate": 1.9816271834476642e-05,
225
+ "loss": 1.5828,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.16,
230
+ "grad_norm": 16.25,
231
+ "learning_rate": 1.9781476007338058e-05,
232
+ "loss": 1.3252,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.165,
237
+ "grad_norm": 15.9375,
238
+ "learning_rate": 1.9743700647852356e-05,
239
+ "loss": 1.3516,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.17,
244
+ "grad_norm": 38.75,
245
+ "learning_rate": 1.9702957262759964e-05,
246
+ "loss": 1.4135,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.175,
251
+ "grad_norm": 64.0,
252
+ "learning_rate": 1.9659258262890683e-05,
253
+ "loss": 1.3307,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.18,
258
+ "grad_norm": 109.0,
259
+ "learning_rate": 1.961261695938319e-05,
260
+ "loss": 1.2195,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.185,
265
+ "grad_norm": 11.3125,
266
+ "learning_rate": 1.9563047559630356e-05,
267
+ "loss": 1.2945,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.19,
272
+ "grad_norm": 11.0625,
273
+ "learning_rate": 1.9510565162951538e-05,
274
+ "loss": 1.1441,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.195,
279
+ "grad_norm": 36.0,
280
+ "learning_rate": 1.945518575599317e-05,
281
+ "loss": 1.3148,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.2,
286
+ "grad_norm": 31.0,
287
+ "learning_rate": 1.9396926207859085e-05,
288
+ "loss": 1.2035,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.205,
293
+ "grad_norm": 81.0,
294
+ "learning_rate": 1.9335804264972018e-05,
295
+ "loss": 1.1465,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.21,
300
+ "grad_norm": 77.0,
301
+ "learning_rate": 1.9271838545667876e-05,
302
+ "loss": 1.1883,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.215,
307
+ "grad_norm": 64.0,
308
+ "learning_rate": 1.9205048534524405e-05,
309
+ "loss": 1.1635,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.22,
314
+ "grad_norm": 12.3125,
315
+ "learning_rate": 1.913545457642601e-05,
316
+ "loss": 0.8765,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.225,
321
+ "grad_norm": 15.8125,
322
+ "learning_rate": 1.9063077870366504e-05,
323
+ "loss": 1.2404,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.23,
328
+ "grad_norm": 9.3125,
329
+ "learning_rate": 1.8987940462991673e-05,
330
+ "loss": 1.082,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.235,
335
+ "grad_norm": 111.0,
336
+ "learning_rate": 1.891006524188368e-05,
337
+ "loss": 0.749,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.24,
342
+ "grad_norm": 49.75,
343
+ "learning_rate": 1.8829475928589272e-05,
344
+ "loss": 1.0076,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.245,
349
+ "grad_norm": 16.625,
350
+ "learning_rate": 1.874619707139396e-05,
351
+ "loss": 0.9004,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.25,
356
+ "grad_norm": 34.75,
357
+ "learning_rate": 1.866025403784439e-05,
358
+ "loss": 0.7954,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.255,
363
+ "grad_norm": 80.0,
364
+ "learning_rate": 1.8571673007021124e-05,
365
+ "loss": 0.9604,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.26,
370
+ "grad_norm": 68.5,
371
+ "learning_rate": 1.848048096156426e-05,
372
+ "loss": 0.6887,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.265,
377
+ "grad_norm": 10.625,
378
+ "learning_rate": 1.8386705679454243e-05,
379
+ "loss": 0.7305,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.27,
384
+ "grad_norm": 7.59375,
385
+ "learning_rate": 1.8290375725550417e-05,
386
+ "loss": 0.6724,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 0.275,
391
+ "grad_norm": 9.0625,
392
+ "learning_rate": 1.819152044288992e-05,
393
+ "loss": 0.7522,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 0.28,
398
+ "grad_norm": 15.8125,
399
+ "learning_rate": 1.8090169943749477e-05,
400
+ "loss": 0.6078,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 0.285,
405
+ "grad_norm": 20.625,
406
+ "learning_rate": 1.798635510047293e-05,
407
+ "loss": 0.6214,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 0.29,
412
+ "grad_norm": 46.5,
413
+ "learning_rate": 1.788010753606722e-05,
414
+ "loss": 0.6955,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 0.295,
419
+ "grad_norm": 6.96875,
420
+ "learning_rate": 1.777145961456971e-05,
421
+ "loss": 0.5002,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 0.3,
426
+ "grad_norm": 9.875,
427
+ "learning_rate": 1.766044443118978e-05,
428
+ "loss": 0.7948,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 0.305,
433
+ "grad_norm": 6.125,
434
+ "learning_rate": 1.7547095802227723e-05,
435
+ "loss": 0.4094,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 0.31,
440
+ "grad_norm": 5.0625,
441
+ "learning_rate": 1.7431448254773943e-05,
442
+ "loss": 0.4871,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 0.315,
447
+ "grad_norm": 6.40625,
448
+ "learning_rate": 1.7313537016191706e-05,
449
+ "loss": 0.6022,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 0.32,
454
+ "grad_norm": 5.59375,
455
+ "learning_rate": 1.7193398003386514e-05,
456
+ "loss": 0.4998,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 0.325,
461
+ "grad_norm": 5.75,
462
+ "learning_rate": 1.7071067811865477e-05,
463
+ "loss": 0.4213,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 0.33,
468
+ "grad_norm": 16.0,
469
+ "learning_rate": 1.6946583704589973e-05,
470
+ "loss": 0.5501,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 0.335,
475
+ "grad_norm": 20.875,
476
+ "learning_rate": 1.6819983600624986e-05,
477
+ "loss": 0.4,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 0.34,
482
+ "grad_norm": 8.0,
483
+ "learning_rate": 1.6691306063588583e-05,
484
+ "loss": 0.3687,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 0.345,
489
+ "grad_norm": 4.375,
490
+ "learning_rate": 1.6560590289905074e-05,
491
+ "loss": 0.4078,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 0.35,
496
+ "grad_norm": 4.78125,
497
+ "learning_rate": 1.6427876096865394e-05,
498
+ "loss": 0.3165,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 0.355,
503
+ "grad_norm": 32.25,
504
+ "learning_rate": 1.6293203910498375e-05,
505
+ "loss": 0.3724,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 0.36,
510
+ "grad_norm": 8.125,
511
+ "learning_rate": 1.6156614753256583e-05,
512
+ "loss": 0.3892,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 0.365,
517
+ "grad_norm": 15.0,
518
+ "learning_rate": 1.6018150231520486e-05,
519
+ "loss": 0.351,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 0.37,
524
+ "grad_norm": 5.46875,
525
+ "learning_rate": 1.5877852522924733e-05,
526
+ "loss": 0.4384,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 0.375,
531
+ "grad_norm": 13.125,
532
+ "learning_rate": 1.573576436351046e-05,
533
+ "loss": 0.346,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 0.38,
538
+ "grad_norm": 3.890625,
539
+ "learning_rate": 1.5591929034707468e-05,
540
+ "loss": 0.303,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 0.385,
545
+ "grad_norm": 15.375,
546
+ "learning_rate": 1.5446390350150272e-05,
547
+ "loss": 0.3484,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 0.39,
552
+ "grad_norm": 6.1875,
553
+ "learning_rate": 1.529919264233205e-05,
554
+ "loss": 0.286,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 0.395,
559
+ "grad_norm": 3.96875,
560
+ "learning_rate": 1.5150380749100545e-05,
561
+ "loss": 0.2787,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 0.4,
566
+ "grad_norm": 4.5625,
567
+ "learning_rate": 1.5000000000000002e-05,
568
+ "loss": 0.2914,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 0.405,
573
+ "grad_norm": 16.125,
574
+ "learning_rate": 1.4848096202463373e-05,
575
+ "loss": 0.2648,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 0.41,
580
+ "grad_norm": 6.0,
581
+ "learning_rate": 1.469471562785891e-05,
582
+ "loss": 0.2437,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 0.415,
587
+ "grad_norm": 5.71875,
588
+ "learning_rate": 1.4539904997395468e-05,
589
+ "loss": 0.2669,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 0.42,
594
+ "grad_norm": 3.5625,
595
+ "learning_rate": 1.4383711467890776e-05,
596
+ "loss": 0.2553,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 0.425,
601
+ "grad_norm": 3.8125,
602
+ "learning_rate": 1.4226182617406996e-05,
603
+ "loss": 0.2304,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 0.43,
608
+ "grad_norm": 6.0625,
609
+ "learning_rate": 1.4067366430758004e-05,
610
+ "loss": 0.2303,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 0.435,
615
+ "grad_norm": 6.0,
616
+ "learning_rate": 1.3907311284892737e-05,
617
+ "loss": 0.2441,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 0.44,
622
+ "grad_norm": 5.71875,
623
+ "learning_rate": 1.3746065934159123e-05,
624
+ "loss": 0.24,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 0.445,
629
+ "grad_norm": 3.546875,
630
+ "learning_rate": 1.3583679495453e-05,
631
+ "loss": 0.2336,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 0.45,
636
+ "grad_norm": 3.140625,
637
+ "learning_rate": 1.342020143325669e-05,
638
+ "loss": 0.2136,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 0.455,
643
+ "grad_norm": 3.71875,
644
+ "learning_rate": 1.3255681544571568e-05,
645
+ "loss": 0.2241,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 0.46,
650
+ "grad_norm": 3.515625,
651
+ "learning_rate": 1.3090169943749475e-05,
652
+ "loss": 0.2322,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 0.465,
657
+ "grad_norm": 4.0625,
658
+ "learning_rate": 1.2923717047227368e-05,
659
+ "loss": 0.2061,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 0.47,
664
+ "grad_norm": 3.8125,
665
+ "learning_rate": 1.2756373558169992e-05,
666
+ "loss": 0.2297,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 0.475,
671
+ "grad_norm": 3.296875,
672
+ "learning_rate": 1.2588190451025209e-05,
673
+ "loss": 0.2035,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 0.48,
678
+ "grad_norm": 3.546875,
679
+ "learning_rate": 1.2419218955996677e-05,
680
+ "loss": 0.2047,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 0.485,
685
+ "grad_norm": 3.203125,
686
+ "learning_rate": 1.2249510543438652e-05,
687
+ "loss": 0.1935,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 0.49,
692
+ "grad_norm": 3.28125,
693
+ "learning_rate": 1.2079116908177592e-05,
694
+ "loss": 0.2045,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 0.495,
699
+ "grad_norm": 2.984375,
700
+ "learning_rate": 1.190808995376545e-05,
701
+ "loss": 0.1985,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 0.5,
706
+ "grad_norm": 2.9375,
707
+ "learning_rate": 1.1736481776669307e-05,
708
+ "loss": 0.1922,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 0.505,
713
+ "grad_norm": 3.140625,
714
+ "learning_rate": 1.156434465040231e-05,
715
+ "loss": 0.1796,
716
+ "step": 1010
717
+ },
718
+ {
719
+ "epoch": 0.51,
720
+ "grad_norm": 3.46875,
721
+ "learning_rate": 1.1391731009600655e-05,
722
+ "loss": 0.1798,
723
+ "step": 1020
724
+ },
725
+ {
726
+ "epoch": 0.515,
727
+ "grad_norm": 2.984375,
728
+ "learning_rate": 1.1218693434051475e-05,
729
+ "loss": 0.173,
730
+ "step": 1030
731
+ },
732
+ {
733
+ "epoch": 0.52,
734
+ "grad_norm": 3.171875,
735
+ "learning_rate": 1.1045284632676535e-05,
736
+ "loss": 0.1625,
737
+ "step": 1040
738
+ },
739
+ {
740
+ "epoch": 0.525,
741
+ "grad_norm": 3.25,
742
+ "learning_rate": 1.0871557427476585e-05,
743
+ "loss": 0.1725,
744
+ "step": 1050
745
+ },
746
+ {
747
+ "epoch": 0.53,
748
+ "grad_norm": 2.421875,
749
+ "learning_rate": 1.0697564737441254e-05,
750
+ "loss": 0.1833,
751
+ "step": 1060
752
+ },
753
+ {
754
+ "epoch": 0.535,
755
+ "grad_norm": 3.078125,
756
+ "learning_rate": 1.0523359562429441e-05,
757
+ "loss": 0.1812,
758
+ "step": 1070
759
+ },
760
+ {
761
+ "epoch": 0.54,
762
+ "grad_norm": 2.765625,
763
+ "learning_rate": 1.0348994967025012e-05,
764
+ "loss": 0.1681,
765
+ "step": 1080
766
+ },
767
+ {
768
+ "epoch": 0.545,
769
+ "grad_norm": 3.046875,
770
+ "learning_rate": 1.0174524064372837e-05,
771
+ "loss": 0.176,
772
+ "step": 1090
773
+ },
774
+ {
775
+ "epoch": 0.55,
776
+ "grad_norm": 3.296875,
777
+ "learning_rate": 1e-05,
778
+ "loss": 0.177,
779
+ "step": 1100
780
+ },
781
+ {
782
+ "epoch": 0.555,
783
+ "grad_norm": 3.703125,
784
+ "learning_rate": 9.825475935627165e-06,
785
+ "loss": 0.1714,
786
+ "step": 1110
787
+ },
788
+ {
789
+ "epoch": 0.56,
790
+ "grad_norm": 2.9375,
791
+ "learning_rate": 9.651005032974994e-06,
792
+ "loss": 0.1514,
793
+ "step": 1120
794
+ },
795
+ {
796
+ "epoch": 0.565,
797
+ "grad_norm": 2.5,
798
+ "learning_rate": 9.476640437570562e-06,
799
+ "loss": 0.1479,
800
+ "step": 1130
801
+ },
802
+ {
803
+ "epoch": 0.57,
804
+ "grad_norm": 3.28125,
805
+ "learning_rate": 9.302435262558748e-06,
806
+ "loss": 0.1638,
807
+ "step": 1140
808
+ },
809
+ {
810
+ "epoch": 0.575,
811
+ "grad_norm": 2.734375,
812
+ "learning_rate": 9.128442572523418e-06,
813
+ "loss": 0.1538,
814
+ "step": 1150
815
+ },
816
+ {
817
+ "epoch": 0.58,
818
+ "grad_norm": 2.65625,
819
+ "learning_rate": 8.954715367323468e-06,
820
+ "loss": 0.1554,
821
+ "step": 1160
822
+ },
823
+ {
824
+ "epoch": 0.585,
825
+ "grad_norm": 2.234375,
826
+ "learning_rate": 8.781306565948528e-06,
827
+ "loss": 0.1489,
828
+ "step": 1170
829
+ },
830
+ {
831
+ "epoch": 0.59,
832
+ "grad_norm": 2.53125,
833
+ "learning_rate": 8.60826899039935e-06,
834
+ "loss": 0.1451,
835
+ "step": 1180
836
+ },
837
+ {
838
+ "epoch": 0.595,
839
+ "grad_norm": 3.65625,
840
+ "learning_rate": 8.43565534959769e-06,
841
+ "loss": 0.1394,
842
+ "step": 1190
843
+ },
844
+ {
845
+ "epoch": 0.6,
846
+ "grad_norm": 2.859375,
847
+ "learning_rate": 8.263518223330698e-06,
848
+ "loss": 0.1447,
849
+ "step": 1200
850
+ },
851
+ {
852
+ "epoch": 0.605,
853
+ "grad_norm": 2.5,
854
+ "learning_rate": 8.091910046234552e-06,
855
+ "loss": 0.1331,
856
+ "step": 1210
857
+ },
858
+ {
859
+ "epoch": 0.61,
860
+ "grad_norm": 1.78125,
861
+ "learning_rate": 7.92088309182241e-06,
862
+ "loss": 0.1393,
863
+ "step": 1220
864
+ },
865
+ {
866
+ "epoch": 0.615,
867
+ "grad_norm": 1.734375,
868
+ "learning_rate": 7.750489456561351e-06,
869
+ "loss": 0.1366,
870
+ "step": 1230
871
+ },
872
+ {
873
+ "epoch": 0.62,
874
+ "grad_norm": 2.234375,
875
+ "learning_rate": 7.580781044003324e-06,
876
+ "loss": 0.1397,
877
+ "step": 1240
878
+ },
879
+ {
880
+ "epoch": 0.625,
881
+ "grad_norm": 2.8125,
882
+ "learning_rate": 7.411809548974792e-06,
883
+ "loss": 0.144,
884
+ "step": 1250
885
+ },
886
+ {
887
+ "epoch": 0.63,
888
+ "grad_norm": 1.75,
889
+ "learning_rate": 7.243626441830009e-06,
890
+ "loss": 0.1424,
891
+ "step": 1260
892
+ },
893
+ {
894
+ "epoch": 0.635,
895
+ "grad_norm": 2.34375,
896
+ "learning_rate": 7.076282952772634e-06,
897
+ "loss": 0.1373,
898
+ "step": 1270
899
+ },
900
+ {
901
+ "epoch": 0.64,
902
+ "grad_norm": 1.5703125,
903
+ "learning_rate": 6.909830056250527e-06,
904
+ "loss": 0.141,
905
+ "step": 1280
906
+ },
907
+ {
908
+ "epoch": 0.645,
909
+ "grad_norm": 2.078125,
910
+ "learning_rate": 6.744318455428436e-06,
911
+ "loss": 0.1408,
912
+ "step": 1290
913
+ },
914
+ {
915
+ "epoch": 0.65,
916
+ "grad_norm": 1.8671875,
917
+ "learning_rate": 6.579798566743314e-06,
918
+ "loss": 0.1423,
919
+ "step": 1300
920
+ },
921
+ {
922
+ "epoch": 0.655,
923
+ "grad_norm": 2.90625,
924
+ "learning_rate": 6.4163205045469975e-06,
925
+ "loss": 0.1425,
926
+ "step": 1310
927
+ },
928
+ {
929
+ "epoch": 1.00325,
930
+ "grad_norm": 2.078125,
931
+ "learning_rate": 6.25393406584088e-06,
932
+ "loss": 0.1304,
933
+ "step": 1320
934
+ },
935
+ {
936
+ "epoch": 1.00825,
937
+ "grad_norm": 1.453125,
938
+ "learning_rate": 6.092688715107265e-06,
939
+ "loss": 0.1265,
940
+ "step": 1330
941
+ },
942
+ {
943
+ "epoch": 1.01325,
944
+ "grad_norm": 2.8125,
945
+ "learning_rate": 5.932633569242e-06,
946
+ "loss": 0.1298,
947
+ "step": 1340
948
+ },
949
+ {
950
+ "epoch": 1.01825,
951
+ "grad_norm": 2.5625,
952
+ "learning_rate": 5.773817382593008e-06,
953
+ "loss": 0.1219,
954
+ "step": 1350
955
+ },
956
+ {
957
+ "epoch": 1.02325,
958
+ "grad_norm": 2.890625,
959
+ "learning_rate": 5.616288532109225e-06,
960
+ "loss": 0.1362,
961
+ "step": 1360
962
+ },
963
+ {
964
+ "epoch": 1.02825,
965
+ "grad_norm": 2.40625,
966
+ "learning_rate": 5.460095002604533e-06,
967
+ "loss": 0.1371,
968
+ "step": 1370
969
+ },
970
+ {
971
+ "epoch": 1.03325,
972
+ "grad_norm": 1.578125,
973
+ "learning_rate": 5.305284372141095e-06,
974
+ "loss": 0.132,
975
+ "step": 1380
976
+ },
977
+ {
978
+ "epoch": 1.03825,
979
+ "grad_norm": 2.421875,
980
+ "learning_rate": 5.151903797536631e-06,
981
+ "loss": 0.1396,
982
+ "step": 1390
983
+ },
984
+ {
985
+ "epoch": 1.04325,
986
+ "grad_norm": 1.9453125,
987
+ "learning_rate": 5.000000000000003e-06,
988
+ "loss": 0.1356,
989
+ "step": 1400
990
+ },
991
+ {
992
+ "epoch": 1.04825,
993
+ "grad_norm": 1.8359375,
994
+ "learning_rate": 4.849619250899458e-06,
995
+ "loss": 0.1284,
996
+ "step": 1410
997
+ },
998
+ {
999
+ "epoch": 1.05325,
1000
+ "grad_norm": 2.03125,
1001
+ "learning_rate": 4.700807357667953e-06,
1002
+ "loss": 0.1302,
1003
+ "step": 1420
1004
+ },
1005
+ {
1006
+ "epoch": 1.05825,
1007
+ "grad_norm": 1.6875,
1008
+ "learning_rate": 4.5536096498497295e-06,
1009
+ "loss": 0.1325,
1010
+ "step": 1430
1011
+ },
1012
+ {
1013
+ "epoch": 1.06325,
1014
+ "grad_norm": 2.3125,
1015
+ "learning_rate": 4.408070965292534e-06,
1016
+ "loss": 0.1279,
1017
+ "step": 1440
1018
+ },
1019
+ {
1020
+ "epoch": 1.06825,
1021
+ "grad_norm": 2.140625,
1022
+ "learning_rate": 4.264235636489542e-06,
1023
+ "loss": 0.1268,
1024
+ "step": 1450
1025
+ },
1026
+ {
1027
+ "epoch": 1.07325,
1028
+ "grad_norm": 2.59375,
1029
+ "learning_rate": 4.12214747707527e-06,
1030
+ "loss": 0.1296,
1031
+ "step": 1460
1032
+ },
1033
+ {
1034
+ "epoch": 1.07825,
1035
+ "grad_norm": 2.484375,
1036
+ "learning_rate": 3.981849768479516e-06,
1037
+ "loss": 0.1263,
1038
+ "step": 1470
1039
+ },
1040
+ {
1041
+ "epoch": 1.08325,
1042
+ "grad_norm": 2.703125,
1043
+ "learning_rate": 3.8433852467434175e-06,
1044
+ "loss": 0.1232,
1045
+ "step": 1480
1046
+ },
1047
+ {
1048
+ "epoch": 1.08825,
1049
+ "grad_norm": 1.6640625,
1050
+ "learning_rate": 3.7067960895016277e-06,
1051
+ "loss": 0.1212,
1052
+ "step": 1490
1053
+ },
1054
+ {
1055
+ "epoch": 1.09325,
1056
+ "grad_norm": 1.8046875,
1057
+ "learning_rate": 3.5721239031346067e-06,
1058
+ "loss": 0.1177,
1059
+ "step": 1500
1060
+ }
1061
+ ],
1062
+ "logging_steps": 10,
1063
+ "max_steps": 2000,
1064
+ "num_input_tokens_seen": 0,
1065
+ "num_train_epochs": 9223372036854775807,
1066
+ "save_steps": 500,
1067
+ "stateful_callbacks": {
1068
+ "TrainerControl": {
1069
+ "args": {
1070
+ "should_epoch_stop": false,
1071
+ "should_evaluate": false,
1072
+ "should_log": false,
1073
+ "should_save": true,
1074
+ "should_training_stop": false
1075
+ },
1076
+ "attributes": {}
1077
+ }
1078
+ },
1079
+ "total_flos": 1.4346932754736742e+17,
1080
+ "train_batch_size": 16,
1081
+ "trial_name": null,
1082
+ "trial_params": null
1083
+ }
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e32038c45d54a6607b9dae189c2a754eb22b9c228b351719f8fba20e106598
3
+ size 5368