Liu-Xiang commited on
Commit
5439c2e
·
verified ·
1 Parent(s): c062f04

Upload folder using huggingface_hub

Browse files
checkpoint-1000/config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bigcode/santacoder",
3
+ "activation_function": "gelu_fast",
4
+ "architectures": [
5
+ "GPT2LMHeadCustomModel"
6
+ ],
7
+ "attention_head_type": "multiquery",
8
+ "attn_pdrop": 0.1,
9
+ "auto_map": {
10
+ "AutoConfig": "bigcode/santacoder--configuration_gpt2_mq.GPT2CustomConfig",
11
+ "AutoModelForCausalLM": "bigcode/santacoder--modeling_gpt2_mq.GPT2LMHeadCustomModel"
12
+ },
13
+ "bos_token_id": 49152,
14
+ "embd_pdrop": 0.1,
15
+ "eos_token_id": 49152,
16
+ "initializer_range": 0.02,
17
+ "layer_norm_epsilon": 1e-05,
18
+ "model_type": "gpt2",
19
+ "n_embd": 2048,
20
+ "n_head": 16,
21
+ "n_inner": 8192,
22
+ "n_layer": 24,
23
+ "n_positions": 2048,
24
+ "reorder_and_upcast_attn": false,
25
+ "resid_pdrop": 0.1,
26
+ "scale_attn_by_inverse_layer_idx": false,
27
+ "scale_attn_weights": true,
28
+ "summary_activation": null,
29
+ "summary_first_dropout": 0.1,
30
+ "summary_proj_to_labels": true,
31
+ "summary_type": "cls_index",
32
+ "summary_use_proj": true,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.39.0",
35
+ "use_cache": false,
36
+ "vocab_size": 49280
37
+ }
checkpoint-1000/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 49152,
4
+ "eos_token_id": 49152,
5
+ "transformers_version": "4.39.0",
6
+ "use_cache": false
7
+ }
checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efd4bbf030c5732aee7f4cabde37afe1ecbb0eeb42b15a34d98c039e86bf62f5
3
+ size 4600250264
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0a31627c2fa92a0d547d053232adfdcda06e1e0fbe20de65aca07c02dd0daa4
3
+ size 5626501
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db57b704a1370e676bdcdf3eaa176189f0082bc330b8d530fdacf7853700fdc1
3
+ size 14575
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae472c841adf3819dda221ed75de8337578682a35bb543a62b8daee29826e9bf
3
+ size 627
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,737 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "grad_norm": 4.737985134124756,
14
+ "learning_rate": 5e-06,
15
+ "loss": 1.5912,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.02,
20
+ "grad_norm": 7.149003982543945,
21
+ "learning_rate": 1e-05,
22
+ "loss": 3.2011,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.03,
27
+ "grad_norm": 6.096139430999756,
28
+ "learning_rate": 1.5e-05,
29
+ "loss": 1.2445,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.04,
34
+ "grad_norm": 6.4711103439331055,
35
+ "learning_rate": 2e-05,
36
+ "loss": 1.4005,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.05,
41
+ "grad_norm": 6.055766582489014,
42
+ "learning_rate": 2.5e-05,
43
+ "loss": 0.9311,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.06,
48
+ "grad_norm": 23.06619644165039,
49
+ "learning_rate": 3e-05,
50
+ "loss": 2.2437,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.07,
55
+ "grad_norm": 9.455103874206543,
56
+ "learning_rate": 3.5e-05,
57
+ "loss": 2.7137,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.08,
62
+ "grad_norm": 8.64001750946045,
63
+ "learning_rate": 4e-05,
64
+ "loss": 2.6398,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.09,
69
+ "grad_norm": 58.382110595703125,
70
+ "learning_rate": 4.5e-05,
71
+ "loss": 2.2014,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.1,
76
+ "grad_norm": 5.963547706604004,
77
+ "learning_rate": 5e-05,
78
+ "loss": 2.0685,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.11,
83
+ "grad_norm": 5.909413814544678,
84
+ "learning_rate": 4.99847706754774e-05,
85
+ "loss": 3.1558,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.12,
90
+ "grad_norm": 4.110403537750244,
91
+ "learning_rate": 4.993910125649561e-05,
92
+ "loss": 2.2536,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.13,
97
+ "grad_norm": 17.61237907409668,
98
+ "learning_rate": 4.9863047384206835e-05,
99
+ "loss": 3.2304,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.14,
104
+ "grad_norm": 42.55618667602539,
105
+ "learning_rate": 4.975670171853926e-05,
106
+ "loss": 1.8168,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.15,
111
+ "grad_norm": 7.033435344696045,
112
+ "learning_rate": 4.962019382530521e-05,
113
+ "loss": 0.7345,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.16,
118
+ "grad_norm": 6.939785480499268,
119
+ "learning_rate": 4.9453690018345144e-05,
120
+ "loss": 1.0468,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.17,
125
+ "grad_norm": 7.7263078689575195,
126
+ "learning_rate": 4.925739315689991e-05,
127
+ "loss": 1.6574,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.18,
132
+ "grad_norm": 14.468883514404297,
133
+ "learning_rate": 4.9031542398457974e-05,
134
+ "loss": 1.6383,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.19,
139
+ "grad_norm": 5.9138946533203125,
140
+ "learning_rate": 4.877641290737884e-05,
141
+ "loss": 0.8235,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.2,
146
+ "grad_norm": 11.40308666229248,
147
+ "learning_rate": 4.849231551964771e-05,
148
+ "loss": 0.489,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.21,
153
+ "grad_norm": 20.220449447631836,
154
+ "learning_rate": 4.817959636416969e-05,
155
+ "loss": 1.4448,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.22,
160
+ "grad_norm": 29.097707748413086,
161
+ "learning_rate": 4.783863644106502e-05,
162
+ "loss": 1.7983,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.23,
167
+ "grad_norm": 16.577442169189453,
168
+ "learning_rate": 4.7469851157479177e-05,
169
+ "loss": 1.8216,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.24,
174
+ "grad_norm": 3.516134262084961,
175
+ "learning_rate": 4.707368982147318e-05,
176
+ "loss": 2.0506,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.25,
181
+ "grad_norm": 3.0684404373168945,
182
+ "learning_rate": 4.665063509461097e-05,
183
+ "loss": 1.258,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.26,
188
+ "grad_norm": 4.914421558380127,
189
+ "learning_rate": 4.620120240391065e-05,
190
+ "loss": 0.3911,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.27,
195
+ "grad_norm": 6.704916954040527,
196
+ "learning_rate": 4.572593931387604e-05,
197
+ "loss": 0.6957,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.28,
202
+ "grad_norm": 5.172149181365967,
203
+ "learning_rate": 4.522542485937369e-05,
204
+ "loss": 1.3392,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.29,
209
+ "grad_norm": 2.9674534797668457,
210
+ "learning_rate": 4.4700268840168045e-05,
211
+ "loss": 0.6846,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.3,
216
+ "grad_norm": 2.472653388977051,
217
+ "learning_rate": 4.415111107797445e-05,
218
+ "loss": 0.2735,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.31,
223
+ "grad_norm": 2.405486583709717,
224
+ "learning_rate": 4.357862063693486e-05,
225
+ "loss": 0.2887,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.32,
230
+ "grad_norm": 2.07614803314209,
231
+ "learning_rate": 4.2983495008466276e-05,
232
+ "loss": 0.3026,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.33,
237
+ "grad_norm": 7.051891803741455,
238
+ "learning_rate": 4.2366459261474933e-05,
239
+ "loss": 0.688,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.34,
244
+ "grad_norm": 3.9787468910217285,
245
+ "learning_rate": 4.172826515897146e-05,
246
+ "loss": 2.5623,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.35,
251
+ "grad_norm": 5.217522621154785,
252
+ "learning_rate": 4.1069690242163484e-05,
253
+ "loss": 1.2036,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.36,
258
+ "grad_norm": 5.582896709442139,
259
+ "learning_rate": 4.039153688314145e-05,
260
+ "loss": 1.3976,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.37,
265
+ "grad_norm": 4.498598098754883,
266
+ "learning_rate": 3.969463130731183e-05,
267
+ "loss": 1.6016,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.38,
272
+ "grad_norm": 3.115513324737549,
273
+ "learning_rate": 3.897982258676867e-05,
274
+ "loss": 0.7646,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.39,
279
+ "grad_norm": 3.488847494125366,
280
+ "learning_rate": 3.824798160583012e-05,
281
+ "loss": 0.5091,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.4,
286
+ "grad_norm": 5.372682094573975,
287
+ "learning_rate": 3.7500000000000003e-05,
288
+ "loss": 1.3678,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.41,
293
+ "grad_norm": 5.9022674560546875,
294
+ "learning_rate": 3.673678906964727e-05,
295
+ "loss": 2.1068,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.42,
300
+ "grad_norm": 3.6478700637817383,
301
+ "learning_rate": 3.5959278669726935e-05,
302
+ "loss": 1.4413,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.43,
307
+ "grad_norm": 7.891127109527588,
308
+ "learning_rate": 3.516841607689501e-05,
309
+ "loss": 1.2757,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.44,
314
+ "grad_norm": 5.28378438949585,
315
+ "learning_rate": 3.436516483539781e-05,
316
+ "loss": 0.7885,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.45,
321
+ "grad_norm": 6.047487735748291,
322
+ "learning_rate": 3.355050358314172e-05,
323
+ "loss": 2.3279,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.46,
328
+ "grad_norm": 2.6575264930725098,
329
+ "learning_rate": 3.272542485937369e-05,
330
+ "loss": 0.5281,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.47,
335
+ "grad_norm": 2.3986458778381348,
336
+ "learning_rate": 3.1890933895424976e-05,
337
+ "loss": 0.2615,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.48,
342
+ "grad_norm": 1.6229203939437866,
343
+ "learning_rate": 3.104804738999169e-05,
344
+ "loss": 0.4922,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.49,
349
+ "grad_norm": 1.5451277494430542,
350
+ "learning_rate": 3.0197792270443982e-05,
351
+ "loss": 0.2118,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.5,
356
+ "grad_norm": 1.311353087425232,
357
+ "learning_rate": 2.9341204441673266e-05,
358
+ "loss": 0.1915,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.5,
363
+ "eval_loss": 1.5571472644805908,
364
+ "eval_runtime": 5999.5519,
365
+ "eval_samples_per_second": 5.286,
366
+ "eval_steps_per_second": 5.286,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 0.51,
371
+ "grad_norm": 1.1946927309036255,
372
+ "learning_rate": 2.8479327524001636e-05,
373
+ "loss": 0.1914,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 0.52,
378
+ "grad_norm": 1.0991920232772827,
379
+ "learning_rate": 2.761321158169134e-05,
380
+ "loss": 0.2036,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 0.53,
385
+ "grad_norm": 1.178324580192566,
386
+ "learning_rate": 2.674391184360313e-05,
387
+ "loss": 0.1903,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 0.54,
392
+ "grad_norm": 1.6252118349075317,
393
+ "learning_rate": 2.587248741756253e-05,
394
+ "loss": 0.1897,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 0.55,
399
+ "grad_norm": 2.4392871856689453,
400
+ "learning_rate": 2.5e-05,
401
+ "loss": 0.2962,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 0.56,
406
+ "grad_norm": 4.543885707855225,
407
+ "learning_rate": 2.4127512582437485e-05,
408
+ "loss": 0.7395,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 0.57,
413
+ "grad_norm": 1.537638783454895,
414
+ "learning_rate": 2.3256088156396868e-05,
415
+ "loss": 0.2917,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 0.58,
420
+ "grad_norm": 1.9524015188217163,
421
+ "learning_rate": 2.238678841830867e-05,
422
+ "loss": 0.2657,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 0.59,
427
+ "grad_norm": 3.946528673171997,
428
+ "learning_rate": 2.1520672475998373e-05,
429
+ "loss": 0.9463,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 0.6,
434
+ "grad_norm": 6.766805171966553,
435
+ "learning_rate": 2.0658795558326743e-05,
436
+ "loss": 2.8137,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 0.61,
441
+ "grad_norm": 5.235180377960205,
442
+ "learning_rate": 1.980220772955602e-05,
443
+ "loss": 1.1135,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 0.62,
448
+ "grad_norm": 11.928383827209473,
449
+ "learning_rate": 1.895195261000831e-05,
450
+ "loss": 1.6387,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 0.63,
455
+ "grad_norm": 1.4929327964782715,
456
+ "learning_rate": 1.8109066104575023e-05,
457
+ "loss": 0.5561,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 0.64,
462
+ "grad_norm": 1.8914227485656738,
463
+ "learning_rate": 1.7274575140626318e-05,
464
+ "loss": 0.2111,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 0.65,
469
+ "grad_norm": 3.6962478160858154,
470
+ "learning_rate": 1.6449496416858284e-05,
471
+ "loss": 0.6296,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 0.66,
476
+ "grad_norm": 4.319825649261475,
477
+ "learning_rate": 1.56348351646022e-05,
478
+ "loss": 1.7621,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 0.67,
483
+ "grad_norm": 5.848764896392822,
484
+ "learning_rate": 1.4831583923104999e-05,
485
+ "loss": 1.7205,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 0.68,
490
+ "grad_norm": 7.215566158294678,
491
+ "learning_rate": 1.4040721330273062e-05,
492
+ "loss": 1.4845,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 0.69,
497
+ "grad_norm": 4.199477195739746,
498
+ "learning_rate": 1.3263210930352737e-05,
499
+ "loss": 2.1168,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 0.7,
504
+ "grad_norm": 6.580512046813965,
505
+ "learning_rate": 1.2500000000000006e-05,
506
+ "loss": 1.7481,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 0.71,
511
+ "grad_norm": 7.199362754821777,
512
+ "learning_rate": 1.175201839416988e-05,
513
+ "loss": 1.6295,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 0.72,
518
+ "grad_norm": 3.5996954441070557,
519
+ "learning_rate": 1.1020177413231334e-05,
520
+ "loss": 1.266,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 0.73,
525
+ "grad_norm": 7.7472968101501465,
526
+ "learning_rate": 1.0305368692688174e-05,
527
+ "loss": 0.5478,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 0.74,
532
+ "grad_norm": 3.7776882648468018,
533
+ "learning_rate": 9.608463116858542e-06,
534
+ "loss": 1.1993,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 0.75,
539
+ "grad_norm": 2.427421808242798,
540
+ "learning_rate": 8.930309757836517e-06,
541
+ "loss": 1.6423,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 0.76,
546
+ "grad_norm": 3.11273193359375,
547
+ "learning_rate": 8.271734841028553e-06,
548
+ "loss": 0.4206,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 0.77,
553
+ "grad_norm": 1.6959909200668335,
554
+ "learning_rate": 7.633540738525066e-06,
555
+ "loss": 0.257,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 0.78,
560
+ "grad_norm": 4.651413440704346,
561
+ "learning_rate": 7.016504991533726e-06,
562
+ "loss": 0.3458,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 0.79,
567
+ "grad_norm": 4.119582653045654,
568
+ "learning_rate": 6.421379363065142e-06,
569
+ "loss": 1.721,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 0.8,
574
+ "grad_norm": 4.266563415527344,
575
+ "learning_rate": 5.848888922025553e-06,
576
+ "loss": 1.2284,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 0.81,
581
+ "grad_norm": 2.50252366065979,
582
+ "learning_rate": 5.299731159831953e-06,
583
+ "loss": 1.1696,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 0.82,
588
+ "grad_norm": 2.9079418182373047,
589
+ "learning_rate": 4.7745751406263165e-06,
590
+ "loss": 0.953,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 0.83,
595
+ "grad_norm": 3.4728331565856934,
596
+ "learning_rate": 4.274060686123959e-06,
597
+ "loss": 1.7072,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 0.84,
602
+ "grad_norm": 1.1848517656326294,
603
+ "learning_rate": 3.798797596089351e-06,
604
+ "loss": 0.1604,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 0.85,
609
+ "grad_norm": 0.647183895111084,
610
+ "learning_rate": 3.3493649053890326e-06,
611
+ "loss": 0.0397,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 0.86,
616
+ "grad_norm": 7.117775917053223,
617
+ "learning_rate": 2.9263101785268254e-06,
618
+ "loss": 0.8525,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 0.87,
623
+ "grad_norm": 4.795046329498291,
624
+ "learning_rate": 2.5301488425208296e-06,
625
+ "loss": 5.167,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 0.88,
630
+ "grad_norm": 4.327600955963135,
631
+ "learning_rate": 2.1613635589349756e-06,
632
+ "loss": 4.8475,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 0.89,
637
+ "grad_norm": 4.947289943695068,
638
+ "learning_rate": 1.8204036358303173e-06,
639
+ "loss": 5.0221,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 0.9,
644
+ "grad_norm": 3.9482979774475098,
645
+ "learning_rate": 1.5076844803522922e-06,
646
+ "loss": 4.9018,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 0.91,
651
+ "grad_norm": 6.357035160064697,
652
+ "learning_rate": 1.2235870926211619e-06,
653
+ "loss": 4.4224,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 0.92,
658
+ "grad_norm": 3.0725436210632324,
659
+ "learning_rate": 9.684576015420278e-07,
660
+ "loss": 0.8727,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 0.93,
665
+ "grad_norm": 2.637556552886963,
666
+ "learning_rate": 7.426068431000882e-07,
667
+ "loss": 0.5169,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 0.94,
672
+ "grad_norm": 2.796225070953369,
673
+ "learning_rate": 5.463099816548579e-07,
674
+ "loss": 0.6463,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 0.95,
679
+ "grad_norm": 2.2664058208465576,
680
+ "learning_rate": 3.7980617469479953e-07,
681
+ "loss": 0.2752,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 0.96,
686
+ "grad_norm": 4.001941680908203,
687
+ "learning_rate": 2.4329828146074095e-07,
688
+ "loss": 0.5726,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 0.97,
693
+ "grad_norm": 3.260882616043091,
694
+ "learning_rate": 1.3695261579316777e-07,
695
+ "loss": 1.4077,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 0.98,
700
+ "grad_norm": 5.255369663238525,
701
+ "learning_rate": 6.089874350439506e-08,
702
+ "loss": 1.4728,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 0.99,
707
+ "grad_norm": 3.699906826019287,
708
+ "learning_rate": 1.522932452260595e-08,
709
+ "loss": 1.2553,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 1.0,
714
+ "grad_norm": 3.7628183364868164,
715
+ "learning_rate": 0.0,
716
+ "loss": 0.6231,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 1.0,
721
+ "eval_loss": 1.401175856590271,
722
+ "eval_runtime": 6003.7558,
723
+ "eval_samples_per_second": 5.282,
724
+ "eval_steps_per_second": 5.282,
725
+ "step": 1000
726
+ }
727
+ ],
728
+ "logging_steps": 10,
729
+ "max_steps": 1000,
730
+ "num_input_tokens_seen": 0,
731
+ "num_train_epochs": 9223372036854775807,
732
+ "save_steps": 500,
733
+ "total_flos": 2.5061788483584e+16,
734
+ "train_batch_size": 1,
735
+ "trial_name": null,
736
+ "trial_params": null
737
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cac715ea29c6ed9a521261a949f9b665c6376f02e17cd6890518ef9424eca7c
3
+ size 4475
checkpoint-500/config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bigcode/santacoder",
3
+ "activation_function": "gelu_fast",
4
+ "architectures": [
5
+ "GPT2LMHeadCustomModel"
6
+ ],
7
+ "attention_head_type": "multiquery",
8
+ "attn_pdrop": 0.1,
9
+ "auto_map": {
10
+ "AutoConfig": "bigcode/santacoder--configuration_gpt2_mq.GPT2CustomConfig",
11
+ "AutoModelForCausalLM": "bigcode/santacoder--modeling_gpt2_mq.GPT2LMHeadCustomModel"
12
+ },
13
+ "bos_token_id": 49152,
14
+ "embd_pdrop": 0.1,
15
+ "eos_token_id": 49152,
16
+ "initializer_range": 0.02,
17
+ "layer_norm_epsilon": 1e-05,
18
+ "model_type": "gpt2",
19
+ "n_embd": 2048,
20
+ "n_head": 16,
21
+ "n_inner": 8192,
22
+ "n_layer": 24,
23
+ "n_positions": 2048,
24
+ "reorder_and_upcast_attn": false,
25
+ "resid_pdrop": 0.1,
26
+ "scale_attn_by_inverse_layer_idx": false,
27
+ "scale_attn_weights": true,
28
+ "summary_activation": null,
29
+ "summary_first_dropout": 0.1,
30
+ "summary_proj_to_labels": true,
31
+ "summary_type": "cls_index",
32
+ "summary_use_proj": true,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.39.0",
35
+ "use_cache": false,
36
+ "vocab_size": 49280
37
+ }
checkpoint-500/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 49152,
4
+ "eos_token_id": 49152,
5
+ "transformers_version": "4.39.0",
6
+ "use_cache": false
7
+ }
checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7f5dfdc73dc9d16178e3cbd2646a5f7375b80500198ce945aae3c1f2a28ef6b
3
+ size 4600250264
checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d02a23d28e8df070afab79c28a37daac3bb73dca97b2ddd367c25b5897580ce6
3
+ size 5626501
checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b72ec93c846029ba8b8c0e0094f0b5e5f1e9048e77d9beb72b51201d519eafff
3
+ size 14575
checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3717949c8d1ecf6bba8ac95e6de55d9545439624ed421eef766cec11d2addb7
3
+ size 627
checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,379 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.5,
5
+ "eval_steps": 500,
6
+ "global_step": 500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "grad_norm": 4.737985134124756,
14
+ "learning_rate": 5e-06,
15
+ "loss": 1.5912,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.02,
20
+ "grad_norm": 7.149003982543945,
21
+ "learning_rate": 1e-05,
22
+ "loss": 3.2011,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.03,
27
+ "grad_norm": 6.096139430999756,
28
+ "learning_rate": 1.5e-05,
29
+ "loss": 1.2445,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.04,
34
+ "grad_norm": 6.4711103439331055,
35
+ "learning_rate": 2e-05,
36
+ "loss": 1.4005,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.05,
41
+ "grad_norm": 6.055766582489014,
42
+ "learning_rate": 2.5e-05,
43
+ "loss": 0.9311,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.06,
48
+ "grad_norm": 23.06619644165039,
49
+ "learning_rate": 3e-05,
50
+ "loss": 2.2437,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.07,
55
+ "grad_norm": 9.455103874206543,
56
+ "learning_rate": 3.5e-05,
57
+ "loss": 2.7137,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.08,
62
+ "grad_norm": 8.64001750946045,
63
+ "learning_rate": 4e-05,
64
+ "loss": 2.6398,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.09,
69
+ "grad_norm": 58.382110595703125,
70
+ "learning_rate": 4.5e-05,
71
+ "loss": 2.2014,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.1,
76
+ "grad_norm": 5.963547706604004,
77
+ "learning_rate": 5e-05,
78
+ "loss": 2.0685,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.11,
83
+ "grad_norm": 5.909413814544678,
84
+ "learning_rate": 4.99847706754774e-05,
85
+ "loss": 3.1558,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.12,
90
+ "grad_norm": 4.110403537750244,
91
+ "learning_rate": 4.993910125649561e-05,
92
+ "loss": 2.2536,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.13,
97
+ "grad_norm": 17.61237907409668,
98
+ "learning_rate": 4.9863047384206835e-05,
99
+ "loss": 3.2304,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.14,
104
+ "grad_norm": 42.55618667602539,
105
+ "learning_rate": 4.975670171853926e-05,
106
+ "loss": 1.8168,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.15,
111
+ "grad_norm": 7.033435344696045,
112
+ "learning_rate": 4.962019382530521e-05,
113
+ "loss": 0.7345,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.16,
118
+ "grad_norm": 6.939785480499268,
119
+ "learning_rate": 4.9453690018345144e-05,
120
+ "loss": 1.0468,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.17,
125
+ "grad_norm": 7.7263078689575195,
126
+ "learning_rate": 4.925739315689991e-05,
127
+ "loss": 1.6574,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.18,
132
+ "grad_norm": 14.468883514404297,
133
+ "learning_rate": 4.9031542398457974e-05,
134
+ "loss": 1.6383,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.19,
139
+ "grad_norm": 5.9138946533203125,
140
+ "learning_rate": 4.877641290737884e-05,
141
+ "loss": 0.8235,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.2,
146
+ "grad_norm": 11.40308666229248,
147
+ "learning_rate": 4.849231551964771e-05,
148
+ "loss": 0.489,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.21,
153
+ "grad_norm": 20.220449447631836,
154
+ "learning_rate": 4.817959636416969e-05,
155
+ "loss": 1.4448,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.22,
160
+ "grad_norm": 29.097707748413086,
161
+ "learning_rate": 4.783863644106502e-05,
162
+ "loss": 1.7983,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.23,
167
+ "grad_norm": 16.577442169189453,
168
+ "learning_rate": 4.7469851157479177e-05,
169
+ "loss": 1.8216,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.24,
174
+ "grad_norm": 3.516134262084961,
175
+ "learning_rate": 4.707368982147318e-05,
176
+ "loss": 2.0506,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.25,
181
+ "grad_norm": 3.0684404373168945,
182
+ "learning_rate": 4.665063509461097e-05,
183
+ "loss": 1.258,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.26,
188
+ "grad_norm": 4.914421558380127,
189
+ "learning_rate": 4.620120240391065e-05,
190
+ "loss": 0.3911,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.27,
195
+ "grad_norm": 6.704916954040527,
196
+ "learning_rate": 4.572593931387604e-05,
197
+ "loss": 0.6957,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.28,
202
+ "grad_norm": 5.172149181365967,
203
+ "learning_rate": 4.522542485937369e-05,
204
+ "loss": 1.3392,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.29,
209
+ "grad_norm": 2.9674534797668457,
210
+ "learning_rate": 4.4700268840168045e-05,
211
+ "loss": 0.6846,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.3,
216
+ "grad_norm": 2.472653388977051,
217
+ "learning_rate": 4.415111107797445e-05,
218
+ "loss": 0.2735,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.31,
223
+ "grad_norm": 2.405486583709717,
224
+ "learning_rate": 4.357862063693486e-05,
225
+ "loss": 0.2887,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.32,
230
+ "grad_norm": 2.07614803314209,
231
+ "learning_rate": 4.2983495008466276e-05,
232
+ "loss": 0.3026,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.33,
237
+ "grad_norm": 7.051891803741455,
238
+ "learning_rate": 4.2366459261474933e-05,
239
+ "loss": 0.688,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.34,
244
+ "grad_norm": 3.9787468910217285,
245
+ "learning_rate": 4.172826515897146e-05,
246
+ "loss": 2.5623,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.35,
251
+ "grad_norm": 5.217522621154785,
252
+ "learning_rate": 4.1069690242163484e-05,
253
+ "loss": 1.2036,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.36,
258
+ "grad_norm": 5.582896709442139,
259
+ "learning_rate": 4.039153688314145e-05,
260
+ "loss": 1.3976,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.37,
265
+ "grad_norm": 4.498598098754883,
266
+ "learning_rate": 3.969463130731183e-05,
267
+ "loss": 1.6016,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.38,
272
+ "grad_norm": 3.115513324737549,
273
+ "learning_rate": 3.897982258676867e-05,
274
+ "loss": 0.7646,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.39,
279
+ "grad_norm": 3.488847494125366,
280
+ "learning_rate": 3.824798160583012e-05,
281
+ "loss": 0.5091,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.4,
286
+ "grad_norm": 5.372682094573975,
287
+ "learning_rate": 3.7500000000000003e-05,
288
+ "loss": 1.3678,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.41,
293
+ "grad_norm": 5.9022674560546875,
294
+ "learning_rate": 3.673678906964727e-05,
295
+ "loss": 2.1068,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.42,
300
+ "grad_norm": 3.6478700637817383,
301
+ "learning_rate": 3.5959278669726935e-05,
302
+ "loss": 1.4413,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.43,
307
+ "grad_norm": 7.891127109527588,
308
+ "learning_rate": 3.516841607689501e-05,
309
+ "loss": 1.2757,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.44,
314
+ "grad_norm": 5.28378438949585,
315
+ "learning_rate": 3.436516483539781e-05,
316
+ "loss": 0.7885,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.45,
321
+ "grad_norm": 6.047487735748291,
322
+ "learning_rate": 3.355050358314172e-05,
323
+ "loss": 2.3279,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.46,
328
+ "grad_norm": 2.6575264930725098,
329
+ "learning_rate": 3.272542485937369e-05,
330
+ "loss": 0.5281,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.47,
335
+ "grad_norm": 2.3986458778381348,
336
+ "learning_rate": 3.1890933895424976e-05,
337
+ "loss": 0.2615,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.48,
342
+ "grad_norm": 1.6229203939437866,
343
+ "learning_rate": 3.104804738999169e-05,
344
+ "loss": 0.4922,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.49,
349
+ "grad_norm": 1.5451277494430542,
350
+ "learning_rate": 3.0197792270443982e-05,
351
+ "loss": 0.2118,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.5,
356
+ "grad_norm": 1.311353087425232,
357
+ "learning_rate": 2.9341204441673266e-05,
358
+ "loss": 0.1915,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.5,
363
+ "eval_loss": 1.5571472644805908,
364
+ "eval_runtime": 5999.5519,
365
+ "eval_samples_per_second": 5.286,
366
+ "eval_steps_per_second": 5.286,
367
+ "step": 500
368
+ }
369
+ ],
370
+ "logging_steps": 10,
371
+ "max_steps": 1000,
372
+ "num_input_tokens_seen": 0,
373
+ "num_train_epochs": 9223372036854775807,
374
+ "save_steps": 500,
375
+ "total_flos": 1.2530894241792e+16,
376
+ "train_batch_size": 1,
377
+ "trial_name": null,
378
+ "trial_params": null
379
+ }
checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cac715ea29c6ed9a521261a949f9b665c6376f02e17cd6890518ef9424eca7c
3
+ size 4475
config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bigcode/santacoder",
3
+ "activation_function": "gelu_fast",
4
+ "architectures": [
5
+ "GPT2LMHeadCustomModel"
6
+ ],
7
+ "attention_head_type": "multiquery",
8
+ "attn_pdrop": 0.1,
9
+ "auto_map": {
10
+ "AutoConfig": "bigcode/santacoder--configuration_gpt2_mq.GPT2CustomConfig",
11
+ "AutoModelForCausalLM": "bigcode/santacoder--modeling_gpt2_mq.GPT2LMHeadCustomModel"
12
+ },
13
+ "bos_token_id": 49152,
14
+ "embd_pdrop": 0.1,
15
+ "eos_token_id": 49152,
16
+ "initializer_range": 0.02,
17
+ "layer_norm_epsilon": 1e-05,
18
+ "model_type": "gpt2",
19
+ "n_embd": 2048,
20
+ "n_head": 16,
21
+ "n_inner": 8192,
22
+ "n_layer": 24,
23
+ "n_positions": 2048,
24
+ "reorder_and_upcast_attn": false,
25
+ "resid_pdrop": 0.1,
26
+ "scale_attn_by_inverse_layer_idx": false,
27
+ "scale_attn_weights": true,
28
+ "summary_activation": null,
29
+ "summary_first_dropout": 0.1,
30
+ "summary_proj_to_labels": true,
31
+ "summary_type": "cls_index",
32
+ "summary_use_proj": true,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.39.0",
35
+ "use_cache": false,
36
+ "vocab_size": 49280
37
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 49152,
4
+ "eos_token_id": 49152,
5
+ "transformers_version": "4.39.0",
6
+ "use_cache": false
7
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efd4bbf030c5732aee7f4cabde37afe1ecbb0eeb42b15a34d98c039e86bf62f5
3
+ size 4600250264
runs/Mar21_13-34-30_llm-collection-workbench-0/events.out.tfevents.1711028071.llm-collection-workbench-0.451.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1ce72a0f849389400dd6a2838f51d899d72a27a6335b7ea6d372414eb91dff7
3
+ size 5051
runs/Mar21_13-43-57_llm-collection-workbench-0/events.out.tfevents.1711028638.llm-collection-workbench-0.253.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:821e0916e75696c8e1751f53b60594282db9977adb9048c8b3419d9457677cfc
3
+ size 5051
runs/Mar21_14-56-33_llm-collection-workbench-0/events.out.tfevents.1711032994.llm-collection-workbench-0.146.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ddadb0cd7797625e32ad33ae3a0f4bb3ac827907ed119dc6e7b0d347e3f310a
3
+ size 6500
runs/Mar21_15-07-18_llm-collection-workbench-0/events.out.tfevents.1711033639.llm-collection-workbench-0.244.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81d4f893b305440317552119564798db1ca853c162ee9e0d4c44b002a76c2468
3
+ size 26999
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cac715ea29c6ed9a521261a949f9b665c6376f02e17cd6890518ef9424eca7c
3
+ size 4475