ray commited on
Commit
f3aab75
1 Parent(s): 3cf95e3
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "saved_models/RedPajama-INCITE-Base-3B-v1",
3
+ "architectures": [
4
+ "GPTNeoXForCausalLM"
5
+ ],
6
+ "bos_token_id": 0,
7
+ "classifier_dropout": 0.1,
8
+ "eos_token_id": 0,
9
+ "hidden_act": "gelu",
10
+ "hidden_size": 2560,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 10240,
13
+ "layer_norm_eps": 1e-05,
14
+ "max_position_embeddings": 4096,
15
+ "model_type": "gpt_neox",
16
+ "num_attention_heads": 32,
17
+ "num_hidden_layers": 32,
18
+ "rotary_emb_base": 10000,
19
+ "rotary_pct": 1.0,
20
+ "tie_word_embeddings": false,
21
+ "torch_dtype": "float16",
22
+ "transformers_version": "4.30.0.dev0",
23
+ "use_cache": true,
24
+ "use_parallel_residual": false,
25
+ "vocab_size": 50288
26
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 0,
5
+ "transformers_version": "4.30.0.dev0"
6
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9db41aa9a2a1cdeee635876e154b532a5c261b596fad931f06090d82787db31b
3
+ size 6087285565
special_tokens_map.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|prompter|>",
4
+ "<|system|>",
5
+ "<|prefix_begin|>",
6
+ "<|prefix_end|>",
7
+ "<|assistant|>"
8
+ ],
9
+ "bos_token": "<|endoftext|>",
10
+ "eos_token": "<|endoftext|>",
11
+ "pad_token": "<|padding|>",
12
+ "sep_token": "<|endoftext|>",
13
+ "unk_token": "<|endoftext|>"
14
+ }
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": "GPTNeoXTokenizer",
8
+ "unk_token": "<|endoftext|>"
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,1252 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.3579935988247023,
5
+ "global_step": 2000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.02,
12
+ "learning_rate": 3.0000000000000004e-08,
13
+ "loss": 1.3803,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 8e-08,
19
+ "loss": 1.3294,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.05,
24
+ "learning_rate": 1.3e-07,
25
+ "loss": 1.2867,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.07,
30
+ "learning_rate": 1.8e-07,
31
+ "loss": 1.2626,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.08,
36
+ "learning_rate": 2.3000000000000002e-07,
37
+ "loss": 1.2373,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.1,
42
+ "learning_rate": 2.8e-07,
43
+ "loss": 1.2286,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "learning_rate": 3.3e-07,
49
+ "loss": 1.2056,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.13,
54
+ "learning_rate": 3.8e-07,
55
+ "loss": 1.2047,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.15,
60
+ "learning_rate": 4.3e-07,
61
+ "loss": 1.1813,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.17,
66
+ "learning_rate": 4.800000000000001e-07,
67
+ "loss": 1.1672,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.18,
72
+ "learning_rate": 5.3e-07,
73
+ "loss": 1.1585,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.2,
78
+ "learning_rate": 5.800000000000001e-07,
79
+ "loss": 1.1429,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.22,
84
+ "learning_rate": 6.3e-07,
85
+ "loss": 1.1384,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.24,
90
+ "learning_rate": 6.800000000000001e-07,
91
+ "loss": 1.1265,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.25,
96
+ "learning_rate": 7.3e-07,
97
+ "loss": 1.1375,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.27,
102
+ "learning_rate": 7.8e-07,
103
+ "loss": 1.1396,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.29,
108
+ "learning_rate": 8.300000000000001e-07,
109
+ "loss": 1.1302,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.3,
114
+ "learning_rate": 8.8e-07,
115
+ "loss": 1.1213,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.32,
120
+ "learning_rate": 9.300000000000001e-07,
121
+ "loss": 1.1183,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.34,
126
+ "learning_rate": 9.800000000000001e-07,
127
+ "loss": 1.1081,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.35,
132
+ "learning_rate": 1.03e-06,
133
+ "loss": 1.0939,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.37,
138
+ "learning_rate": 1.08e-06,
139
+ "loss": 1.0826,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.39,
144
+ "learning_rate": 1.1300000000000002e-06,
145
+ "loss": 1.1104,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.4,
150
+ "learning_rate": 1.1800000000000001e-06,
151
+ "loss": 1.111,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.42,
156
+ "learning_rate": 1.23e-06,
157
+ "loss": 1.0862,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.44,
162
+ "learning_rate": 1.28e-06,
163
+ "loss": 1.0783,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.45,
168
+ "learning_rate": 1.3300000000000002e-06,
169
+ "loss": 1.0762,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.47,
174
+ "learning_rate": 1.3800000000000001e-06,
175
+ "loss": 1.0928,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.49,
180
+ "learning_rate": 1.43e-06,
181
+ "loss": 1.0884,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.5,
186
+ "learning_rate": 1.48e-06,
187
+ "loss": 1.068,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.52,
192
+ "learning_rate": 1.5300000000000002e-06,
193
+ "loss": 1.0785,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.54,
198
+ "learning_rate": 1.5800000000000001e-06,
199
+ "loss": 1.0538,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.55,
204
+ "learning_rate": 1.6300000000000003e-06,
205
+ "loss": 1.0734,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.57,
210
+ "learning_rate": 1.6800000000000002e-06,
211
+ "loss": 1.0642,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.59,
216
+ "learning_rate": 1.73e-06,
217
+ "loss": 1.067,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.6,
222
+ "learning_rate": 1.7800000000000001e-06,
223
+ "loss": 1.079,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.62,
228
+ "learning_rate": 1.83e-06,
229
+ "loss": 1.048,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.64,
234
+ "learning_rate": 1.8800000000000002e-06,
235
+ "loss": 1.0571,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.65,
240
+ "learning_rate": 1.93e-06,
241
+ "loss": 1.0617,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.67,
246
+ "learning_rate": 1.98e-06,
247
+ "loss": 1.0644,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.69,
252
+ "learning_rate": 2.0300000000000005e-06,
253
+ "loss": 1.0569,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.71,
258
+ "learning_rate": 2.08e-06,
259
+ "loss": 1.0429,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.72,
264
+ "learning_rate": 2.13e-06,
265
+ "loss": 1.03,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.74,
270
+ "learning_rate": 2.1800000000000003e-06,
271
+ "loss": 1.0449,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.76,
276
+ "learning_rate": 2.2300000000000002e-06,
277
+ "loss": 1.0462,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.77,
282
+ "learning_rate": 2.28e-06,
283
+ "loss": 1.0272,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.79,
288
+ "learning_rate": 2.33e-06,
289
+ "loss": 1.0221,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.81,
294
+ "learning_rate": 2.38e-06,
295
+ "loss": 1.0307,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.82,
300
+ "learning_rate": 2.43e-06,
301
+ "loss": 1.048,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.84,
306
+ "learning_rate": 2.4800000000000004e-06,
307
+ "loss": 1.0376,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.84,
312
+ "eval_code_evol_accuracy": 0.7292901919811937,
313
+ "eval_code_evol_loss": 1.0595703125,
314
+ "eval_code_evol_runtime": 44.6855,
315
+ "eval_code_evol_samples_per_second": 17.254,
316
+ "eval_code_evol_steps_per_second": 2.171,
317
+ "step": 500
318
+ },
319
+ {
320
+ "epoch": 0.86,
321
+ "learning_rate": 2.5300000000000003e-06,
322
+ "loss": 1.0246,
323
+ "step": 510
324
+ },
325
+ {
326
+ "epoch": 0.87,
327
+ "learning_rate": 2.5800000000000003e-06,
328
+ "loss": 1.0309,
329
+ "step": 520
330
+ },
331
+ {
332
+ "epoch": 0.89,
333
+ "learning_rate": 2.6300000000000002e-06,
334
+ "loss": 1.0391,
335
+ "step": 530
336
+ },
337
+ {
338
+ "epoch": 0.91,
339
+ "learning_rate": 2.68e-06,
340
+ "loss": 1.0193,
341
+ "step": 540
342
+ },
343
+ {
344
+ "epoch": 0.92,
345
+ "learning_rate": 2.7300000000000005e-06,
346
+ "loss": 1.0039,
347
+ "step": 550
348
+ },
349
+ {
350
+ "epoch": 0.94,
351
+ "learning_rate": 2.7800000000000005e-06,
352
+ "loss": 1.0271,
353
+ "step": 560
354
+ },
355
+ {
356
+ "epoch": 0.96,
357
+ "learning_rate": 2.83e-06,
358
+ "loss": 1.0269,
359
+ "step": 570
360
+ },
361
+ {
362
+ "epoch": 0.97,
363
+ "learning_rate": 2.88e-06,
364
+ "loss": 1.0103,
365
+ "step": 580
366
+ },
367
+ {
368
+ "epoch": 0.99,
369
+ "learning_rate": 2.93e-06,
370
+ "loss": 1.0012,
371
+ "step": 590
372
+ },
373
+ {
374
+ "epoch": 1.01,
375
+ "learning_rate": 2.9800000000000003e-06,
376
+ "loss": 1.0011,
377
+ "step": 600
378
+ },
379
+ {
380
+ "epoch": 1.02,
381
+ "learning_rate": 3.0300000000000002e-06,
382
+ "loss": 0.9842,
383
+ "step": 610
384
+ },
385
+ {
386
+ "epoch": 1.04,
387
+ "learning_rate": 3.08e-06,
388
+ "loss": 0.9918,
389
+ "step": 620
390
+ },
391
+ {
392
+ "epoch": 1.06,
393
+ "learning_rate": 3.13e-06,
394
+ "loss": 0.985,
395
+ "step": 630
396
+ },
397
+ {
398
+ "epoch": 1.07,
399
+ "learning_rate": 3.1800000000000005e-06,
400
+ "loss": 0.988,
401
+ "step": 640
402
+ },
403
+ {
404
+ "epoch": 1.09,
405
+ "learning_rate": 3.2300000000000004e-06,
406
+ "loss": 0.9832,
407
+ "step": 650
408
+ },
409
+ {
410
+ "epoch": 1.11,
411
+ "learning_rate": 3.2800000000000004e-06,
412
+ "loss": 0.97,
413
+ "step": 660
414
+ },
415
+ {
416
+ "epoch": 1.12,
417
+ "learning_rate": 3.3300000000000003e-06,
418
+ "loss": 0.9699,
419
+ "step": 670
420
+ },
421
+ {
422
+ "epoch": 1.14,
423
+ "learning_rate": 3.3800000000000007e-06,
424
+ "loss": 0.983,
425
+ "step": 680
426
+ },
427
+ {
428
+ "epoch": 1.16,
429
+ "learning_rate": 3.4300000000000006e-06,
430
+ "loss": 0.9789,
431
+ "step": 690
432
+ },
433
+ {
434
+ "epoch": 1.18,
435
+ "learning_rate": 3.48e-06,
436
+ "loss": 0.9783,
437
+ "step": 700
438
+ },
439
+ {
440
+ "epoch": 1.19,
441
+ "learning_rate": 3.53e-06,
442
+ "loss": 0.9805,
443
+ "step": 710
444
+ },
445
+ {
446
+ "epoch": 1.21,
447
+ "learning_rate": 3.58e-06,
448
+ "loss": 0.9764,
449
+ "step": 720
450
+ },
451
+ {
452
+ "epoch": 1.23,
453
+ "learning_rate": 3.6300000000000004e-06,
454
+ "loss": 0.9754,
455
+ "step": 730
456
+ },
457
+ {
458
+ "epoch": 1.24,
459
+ "learning_rate": 3.6800000000000003e-06,
460
+ "loss": 0.9765,
461
+ "step": 740
462
+ },
463
+ {
464
+ "epoch": 1.26,
465
+ "learning_rate": 3.7300000000000003e-06,
466
+ "loss": 0.9637,
467
+ "step": 750
468
+ },
469
+ {
470
+ "epoch": 1.28,
471
+ "learning_rate": 3.7800000000000002e-06,
472
+ "loss": 0.9689,
473
+ "step": 760
474
+ },
475
+ {
476
+ "epoch": 1.29,
477
+ "learning_rate": 3.830000000000001e-06,
478
+ "loss": 0.9847,
479
+ "step": 770
480
+ },
481
+ {
482
+ "epoch": 1.31,
483
+ "learning_rate": 3.88e-06,
484
+ "loss": 0.9646,
485
+ "step": 780
486
+ },
487
+ {
488
+ "epoch": 1.33,
489
+ "learning_rate": 3.9300000000000005e-06,
490
+ "loss": 0.9844,
491
+ "step": 790
492
+ },
493
+ {
494
+ "epoch": 1.34,
495
+ "learning_rate": 3.980000000000001e-06,
496
+ "loss": 0.9678,
497
+ "step": 800
498
+ },
499
+ {
500
+ "epoch": 1.36,
501
+ "learning_rate": 4.03e-06,
502
+ "loss": 0.9597,
503
+ "step": 810
504
+ },
505
+ {
506
+ "epoch": 1.38,
507
+ "learning_rate": 4.08e-06,
508
+ "loss": 0.981,
509
+ "step": 820
510
+ },
511
+ {
512
+ "epoch": 1.39,
513
+ "learning_rate": 4.13e-06,
514
+ "loss": 0.9736,
515
+ "step": 830
516
+ },
517
+ {
518
+ "epoch": 1.41,
519
+ "learning_rate": 4.18e-06,
520
+ "loss": 0.9702,
521
+ "step": 840
522
+ },
523
+ {
524
+ "epoch": 1.43,
525
+ "learning_rate": 4.23e-06,
526
+ "loss": 0.9676,
527
+ "step": 850
528
+ },
529
+ {
530
+ "epoch": 1.44,
531
+ "learning_rate": 4.2800000000000005e-06,
532
+ "loss": 0.975,
533
+ "step": 860
534
+ },
535
+ {
536
+ "epoch": 1.46,
537
+ "learning_rate": 4.33e-06,
538
+ "loss": 0.961,
539
+ "step": 870
540
+ },
541
+ {
542
+ "epoch": 1.48,
543
+ "learning_rate": 4.38e-06,
544
+ "loss": 0.9572,
545
+ "step": 880
546
+ },
547
+ {
548
+ "epoch": 1.49,
549
+ "learning_rate": 4.430000000000001e-06,
550
+ "loss": 0.9637,
551
+ "step": 890
552
+ },
553
+ {
554
+ "epoch": 1.51,
555
+ "learning_rate": 4.48e-06,
556
+ "loss": 0.9625,
557
+ "step": 900
558
+ },
559
+ {
560
+ "epoch": 1.53,
561
+ "learning_rate": 4.530000000000001e-06,
562
+ "loss": 0.9634,
563
+ "step": 910
564
+ },
565
+ {
566
+ "epoch": 1.54,
567
+ "learning_rate": 4.58e-06,
568
+ "loss": 0.9524,
569
+ "step": 920
570
+ },
571
+ {
572
+ "epoch": 1.56,
573
+ "learning_rate": 4.6300000000000006e-06,
574
+ "loss": 0.957,
575
+ "step": 930
576
+ },
577
+ {
578
+ "epoch": 1.58,
579
+ "learning_rate": 4.680000000000001e-06,
580
+ "loss": 0.9609,
581
+ "step": 940
582
+ },
583
+ {
584
+ "epoch": 1.6,
585
+ "learning_rate": 4.7300000000000005e-06,
586
+ "loss": 0.9684,
587
+ "step": 950
588
+ },
589
+ {
590
+ "epoch": 1.61,
591
+ "learning_rate": 4.78e-06,
592
+ "loss": 0.9607,
593
+ "step": 960
594
+ },
595
+ {
596
+ "epoch": 1.63,
597
+ "learning_rate": 4.83e-06,
598
+ "loss": 0.9577,
599
+ "step": 970
600
+ },
601
+ {
602
+ "epoch": 1.65,
603
+ "learning_rate": 4.880000000000001e-06,
604
+ "loss": 0.9614,
605
+ "step": 980
606
+ },
607
+ {
608
+ "epoch": 1.66,
609
+ "learning_rate": 4.93e-06,
610
+ "loss": 0.953,
611
+ "step": 990
612
+ },
613
+ {
614
+ "epoch": 1.68,
615
+ "learning_rate": 4.980000000000001e-06,
616
+ "loss": 0.9679,
617
+ "step": 1000
618
+ },
619
+ {
620
+ "epoch": 1.68,
621
+ "eval_code_evol_accuracy": 0.7400189369204649,
622
+ "eval_code_evol_loss": 1.001953125,
623
+ "eval_code_evol_runtime": 44.1999,
624
+ "eval_code_evol_samples_per_second": 17.443,
625
+ "eval_code_evol_steps_per_second": 2.195,
626
+ "step": 1000
627
+ },
628
+ {
629
+ "epoch": 1.7,
630
+ "learning_rate": 5.03e-06,
631
+ "loss": 0.949,
632
+ "step": 1010
633
+ },
634
+ {
635
+ "epoch": 1.71,
636
+ "learning_rate": 5.0800000000000005e-06,
637
+ "loss": 0.9567,
638
+ "step": 1020
639
+ },
640
+ {
641
+ "epoch": 1.73,
642
+ "learning_rate": 5.130000000000001e-06,
643
+ "loss": 0.9527,
644
+ "step": 1030
645
+ },
646
+ {
647
+ "epoch": 1.75,
648
+ "learning_rate": 5.18e-06,
649
+ "loss": 0.9614,
650
+ "step": 1040
651
+ },
652
+ {
653
+ "epoch": 1.76,
654
+ "learning_rate": 5.230000000000001e-06,
655
+ "loss": 0.9595,
656
+ "step": 1050
657
+ },
658
+ {
659
+ "epoch": 1.78,
660
+ "learning_rate": 5.28e-06,
661
+ "loss": 0.9482,
662
+ "step": 1060
663
+ },
664
+ {
665
+ "epoch": 1.8,
666
+ "learning_rate": 5.330000000000001e-06,
667
+ "loss": 0.9403,
668
+ "step": 1070
669
+ },
670
+ {
671
+ "epoch": 1.81,
672
+ "learning_rate": 5.380000000000001e-06,
673
+ "loss": 0.9386,
674
+ "step": 1080
675
+ },
676
+ {
677
+ "epoch": 1.83,
678
+ "learning_rate": 5.4300000000000005e-06,
679
+ "loss": 0.9663,
680
+ "step": 1090
681
+ },
682
+ {
683
+ "epoch": 1.85,
684
+ "learning_rate": 5.480000000000001e-06,
685
+ "loss": 0.9491,
686
+ "step": 1100
687
+ },
688
+ {
689
+ "epoch": 1.86,
690
+ "learning_rate": 5.530000000000001e-06,
691
+ "loss": 0.9638,
692
+ "step": 1110
693
+ },
694
+ {
695
+ "epoch": 1.88,
696
+ "learning_rate": 5.580000000000001e-06,
697
+ "loss": 0.9728,
698
+ "step": 1120
699
+ },
700
+ {
701
+ "epoch": 1.9,
702
+ "learning_rate": 5.63e-06,
703
+ "loss": 0.9551,
704
+ "step": 1130
705
+ },
706
+ {
707
+ "epoch": 1.91,
708
+ "learning_rate": 5.68e-06,
709
+ "loss": 0.9432,
710
+ "step": 1140
711
+ },
712
+ {
713
+ "epoch": 1.93,
714
+ "learning_rate": 5.73e-06,
715
+ "loss": 0.9517,
716
+ "step": 1150
717
+ },
718
+ {
719
+ "epoch": 1.95,
720
+ "learning_rate": 5.78e-06,
721
+ "loss": 0.9448,
722
+ "step": 1160
723
+ },
724
+ {
725
+ "epoch": 1.96,
726
+ "learning_rate": 5.83e-06,
727
+ "loss": 0.954,
728
+ "step": 1170
729
+ },
730
+ {
731
+ "epoch": 1.98,
732
+ "learning_rate": 5.8800000000000005e-06,
733
+ "loss": 0.9402,
734
+ "step": 1180
735
+ },
736
+ {
737
+ "epoch": 2.0,
738
+ "learning_rate": 5.93e-06,
739
+ "loss": 0.9257,
740
+ "step": 1190
741
+ },
742
+ {
743
+ "epoch": 2.01,
744
+ "learning_rate": 5.98e-06,
745
+ "loss": 0.8889,
746
+ "step": 1200
747
+ },
748
+ {
749
+ "epoch": 2.03,
750
+ "learning_rate": 6.030000000000001e-06,
751
+ "loss": 0.8903,
752
+ "step": 1210
753
+ },
754
+ {
755
+ "epoch": 2.05,
756
+ "learning_rate": 6.08e-06,
757
+ "loss": 0.8778,
758
+ "step": 1220
759
+ },
760
+ {
761
+ "epoch": 2.07,
762
+ "learning_rate": 6.130000000000001e-06,
763
+ "loss": 0.8759,
764
+ "step": 1230
765
+ },
766
+ {
767
+ "epoch": 2.08,
768
+ "learning_rate": 6.18e-06,
769
+ "loss": 0.8957,
770
+ "step": 1240
771
+ },
772
+ {
773
+ "epoch": 2.1,
774
+ "learning_rate": 6.2300000000000005e-06,
775
+ "loss": 0.8805,
776
+ "step": 1250
777
+ },
778
+ {
779
+ "epoch": 2.12,
780
+ "learning_rate": 6.280000000000001e-06,
781
+ "loss": 0.882,
782
+ "step": 1260
783
+ },
784
+ {
785
+ "epoch": 2.13,
786
+ "learning_rate": 6.33e-06,
787
+ "loss": 0.8887,
788
+ "step": 1270
789
+ },
790
+ {
791
+ "epoch": 2.15,
792
+ "learning_rate": 6.380000000000001e-06,
793
+ "loss": 0.8692,
794
+ "step": 1280
795
+ },
796
+ {
797
+ "epoch": 2.17,
798
+ "learning_rate": 6.43e-06,
799
+ "loss": 0.896,
800
+ "step": 1290
801
+ },
802
+ {
803
+ "epoch": 2.18,
804
+ "learning_rate": 6.480000000000001e-06,
805
+ "loss": 0.8724,
806
+ "step": 1300
807
+ },
808
+ {
809
+ "epoch": 2.2,
810
+ "learning_rate": 6.530000000000001e-06,
811
+ "loss": 0.8705,
812
+ "step": 1310
813
+ },
814
+ {
815
+ "epoch": 2.22,
816
+ "learning_rate": 6.5800000000000005e-06,
817
+ "loss": 0.8805,
818
+ "step": 1320
819
+ },
820
+ {
821
+ "epoch": 2.23,
822
+ "learning_rate": 6.630000000000001e-06,
823
+ "loss": 0.8804,
824
+ "step": 1330
825
+ },
826
+ {
827
+ "epoch": 2.25,
828
+ "learning_rate": 6.680000000000001e-06,
829
+ "loss": 0.8864,
830
+ "step": 1340
831
+ },
832
+ {
833
+ "epoch": 2.27,
834
+ "learning_rate": 6.730000000000001e-06,
835
+ "loss": 0.8908,
836
+ "step": 1350
837
+ },
838
+ {
839
+ "epoch": 2.28,
840
+ "learning_rate": 6.780000000000001e-06,
841
+ "loss": 0.8836,
842
+ "step": 1360
843
+ },
844
+ {
845
+ "epoch": 2.3,
846
+ "learning_rate": 6.830000000000001e-06,
847
+ "loss": 0.8817,
848
+ "step": 1370
849
+ },
850
+ {
851
+ "epoch": 2.32,
852
+ "learning_rate": 6.88e-06,
853
+ "loss": 0.8825,
854
+ "step": 1380
855
+ },
856
+ {
857
+ "epoch": 2.33,
858
+ "learning_rate": 6.93e-06,
859
+ "loss": 0.8928,
860
+ "step": 1390
861
+ },
862
+ {
863
+ "epoch": 2.35,
864
+ "learning_rate": 6.98e-06,
865
+ "loss": 0.885,
866
+ "step": 1400
867
+ },
868
+ {
869
+ "epoch": 2.37,
870
+ "learning_rate": 7.0300000000000005e-06,
871
+ "loss": 0.8756,
872
+ "step": 1410
873
+ },
874
+ {
875
+ "epoch": 2.38,
876
+ "learning_rate": 7.08e-06,
877
+ "loss": 0.8945,
878
+ "step": 1420
879
+ },
880
+ {
881
+ "epoch": 2.4,
882
+ "learning_rate": 7.13e-06,
883
+ "loss": 0.8712,
884
+ "step": 1430
885
+ },
886
+ {
887
+ "epoch": 2.42,
888
+ "learning_rate": 7.180000000000001e-06,
889
+ "loss": 0.8992,
890
+ "step": 1440
891
+ },
892
+ {
893
+ "epoch": 2.43,
894
+ "learning_rate": 7.23e-06,
895
+ "loss": 0.8687,
896
+ "step": 1450
897
+ },
898
+ {
899
+ "epoch": 2.45,
900
+ "learning_rate": 7.280000000000001e-06,
901
+ "loss": 0.8896,
902
+ "step": 1460
903
+ },
904
+ {
905
+ "epoch": 2.47,
906
+ "learning_rate": 7.33e-06,
907
+ "loss": 0.8853,
908
+ "step": 1470
909
+ },
910
+ {
911
+ "epoch": 2.48,
912
+ "learning_rate": 7.3800000000000005e-06,
913
+ "loss": 0.866,
914
+ "step": 1480
915
+ },
916
+ {
917
+ "epoch": 2.5,
918
+ "learning_rate": 7.430000000000001e-06,
919
+ "loss": 0.8834,
920
+ "step": 1490
921
+ },
922
+ {
923
+ "epoch": 2.52,
924
+ "learning_rate": 7.48e-06,
925
+ "loss": 0.8657,
926
+ "step": 1500
927
+ },
928
+ {
929
+ "epoch": 2.52,
930
+ "eval_code_evol_accuracy": 0.7452004701580254,
931
+ "eval_code_evol_loss": 0.98095703125,
932
+ "eval_code_evol_runtime": 43.1875,
933
+ "eval_code_evol_samples_per_second": 17.852,
934
+ "eval_code_evol_steps_per_second": 2.246,
935
+ "step": 1500
936
+ },
937
+ {
938
+ "epoch": 2.54,
939
+ "learning_rate": 7.530000000000001e-06,
940
+ "loss": 0.8986,
941
+ "step": 1510
942
+ },
943
+ {
944
+ "epoch": 2.55,
945
+ "learning_rate": 7.58e-06,
946
+ "loss": 0.8768,
947
+ "step": 1520
948
+ },
949
+ {
950
+ "epoch": 2.57,
951
+ "learning_rate": 7.630000000000001e-06,
952
+ "loss": 0.8873,
953
+ "step": 1530
954
+ },
955
+ {
956
+ "epoch": 2.59,
957
+ "learning_rate": 7.680000000000001e-06,
958
+ "loss": 0.8836,
959
+ "step": 1540
960
+ },
961
+ {
962
+ "epoch": 2.6,
963
+ "learning_rate": 7.73e-06,
964
+ "loss": 0.8689,
965
+ "step": 1550
966
+ },
967
+ {
968
+ "epoch": 2.62,
969
+ "learning_rate": 7.78e-06,
970
+ "loss": 0.8896,
971
+ "step": 1560
972
+ },
973
+ {
974
+ "epoch": 2.64,
975
+ "learning_rate": 7.830000000000001e-06,
976
+ "loss": 0.8916,
977
+ "step": 1570
978
+ },
979
+ {
980
+ "epoch": 2.65,
981
+ "learning_rate": 7.88e-06,
982
+ "loss": 0.8654,
983
+ "step": 1580
984
+ },
985
+ {
986
+ "epoch": 2.67,
987
+ "learning_rate": 7.93e-06,
988
+ "loss": 0.8742,
989
+ "step": 1590
990
+ },
991
+ {
992
+ "epoch": 2.69,
993
+ "learning_rate": 7.980000000000002e-06,
994
+ "loss": 0.8822,
995
+ "step": 1600
996
+ },
997
+ {
998
+ "epoch": 2.7,
999
+ "learning_rate": 8.030000000000001e-06,
1000
+ "loss": 0.8699,
1001
+ "step": 1610
1002
+ },
1003
+ {
1004
+ "epoch": 2.72,
1005
+ "learning_rate": 8.08e-06,
1006
+ "loss": 0.8759,
1007
+ "step": 1620
1008
+ },
1009
+ {
1010
+ "epoch": 2.74,
1011
+ "learning_rate": 8.13e-06,
1012
+ "loss": 0.879,
1013
+ "step": 1630
1014
+ },
1015
+ {
1016
+ "epoch": 2.75,
1017
+ "learning_rate": 8.18e-06,
1018
+ "loss": 0.8819,
1019
+ "step": 1640
1020
+ },
1021
+ {
1022
+ "epoch": 2.77,
1023
+ "learning_rate": 8.23e-06,
1024
+ "loss": 0.8797,
1025
+ "step": 1650
1026
+ },
1027
+ {
1028
+ "epoch": 2.79,
1029
+ "learning_rate": 8.28e-06,
1030
+ "loss": 0.8706,
1031
+ "step": 1660
1032
+ },
1033
+ {
1034
+ "epoch": 2.8,
1035
+ "learning_rate": 8.33e-06,
1036
+ "loss": 0.8732,
1037
+ "step": 1670
1038
+ },
1039
+ {
1040
+ "epoch": 2.82,
1041
+ "learning_rate": 8.380000000000001e-06,
1042
+ "loss": 0.8829,
1043
+ "step": 1680
1044
+ },
1045
+ {
1046
+ "epoch": 2.84,
1047
+ "learning_rate": 8.43e-06,
1048
+ "loss": 0.8914,
1049
+ "step": 1690
1050
+ },
1051
+ {
1052
+ "epoch": 2.85,
1053
+ "learning_rate": 8.48e-06,
1054
+ "loss": 0.8692,
1055
+ "step": 1700
1056
+ },
1057
+ {
1058
+ "epoch": 2.87,
1059
+ "learning_rate": 8.530000000000001e-06,
1060
+ "loss": 0.8975,
1061
+ "step": 1710
1062
+ },
1063
+ {
1064
+ "epoch": 2.89,
1065
+ "learning_rate": 8.580000000000001e-06,
1066
+ "loss": 0.8762,
1067
+ "step": 1720
1068
+ },
1069
+ {
1070
+ "epoch": 2.9,
1071
+ "learning_rate": 8.63e-06,
1072
+ "loss": 0.8955,
1073
+ "step": 1730
1074
+ },
1075
+ {
1076
+ "epoch": 2.92,
1077
+ "learning_rate": 8.68e-06,
1078
+ "loss": 0.8853,
1079
+ "step": 1740
1080
+ },
1081
+ {
1082
+ "epoch": 2.94,
1083
+ "learning_rate": 8.730000000000001e-06,
1084
+ "loss": 0.8702,
1085
+ "step": 1750
1086
+ },
1087
+ {
1088
+ "epoch": 2.96,
1089
+ "learning_rate": 8.78e-06,
1090
+ "loss": 0.8824,
1091
+ "step": 1760
1092
+ },
1093
+ {
1094
+ "epoch": 2.97,
1095
+ "learning_rate": 8.83e-06,
1096
+ "loss": 0.8839,
1097
+ "step": 1770
1098
+ },
1099
+ {
1100
+ "epoch": 2.99,
1101
+ "learning_rate": 8.880000000000001e-06,
1102
+ "loss": 0.885,
1103
+ "step": 1780
1104
+ },
1105
+ {
1106
+ "epoch": 3.01,
1107
+ "learning_rate": 8.930000000000001e-06,
1108
+ "loss": 0.8535,
1109
+ "step": 1790
1110
+ },
1111
+ {
1112
+ "epoch": 3.02,
1113
+ "learning_rate": 8.98e-06,
1114
+ "loss": 0.7718,
1115
+ "step": 1800
1116
+ },
1117
+ {
1118
+ "epoch": 3.04,
1119
+ "learning_rate": 9.030000000000002e-06,
1120
+ "loss": 0.7874,
1121
+ "step": 1810
1122
+ },
1123
+ {
1124
+ "epoch": 3.06,
1125
+ "learning_rate": 9.080000000000001e-06,
1126
+ "loss": 0.7935,
1127
+ "step": 1820
1128
+ },
1129
+ {
1130
+ "epoch": 3.07,
1131
+ "learning_rate": 9.13e-06,
1132
+ "loss": 0.7874,
1133
+ "step": 1830
1134
+ },
1135
+ {
1136
+ "epoch": 3.09,
1137
+ "learning_rate": 9.180000000000002e-06,
1138
+ "loss": 0.7963,
1139
+ "step": 1840
1140
+ },
1141
+ {
1142
+ "epoch": 3.11,
1143
+ "learning_rate": 9.230000000000001e-06,
1144
+ "loss": 0.7788,
1145
+ "step": 1850
1146
+ },
1147
+ {
1148
+ "epoch": 3.12,
1149
+ "learning_rate": 9.280000000000001e-06,
1150
+ "loss": 0.7856,
1151
+ "step": 1860
1152
+ },
1153
+ {
1154
+ "epoch": 3.14,
1155
+ "learning_rate": 9.33e-06,
1156
+ "loss": 0.7824,
1157
+ "step": 1870
1158
+ },
1159
+ {
1160
+ "epoch": 3.16,
1161
+ "learning_rate": 9.38e-06,
1162
+ "loss": 0.7942,
1163
+ "step": 1880
1164
+ },
1165
+ {
1166
+ "epoch": 3.17,
1167
+ "learning_rate": 9.43e-06,
1168
+ "loss": 0.7846,
1169
+ "step": 1890
1170
+ },
1171
+ {
1172
+ "epoch": 3.19,
1173
+ "learning_rate": 9.48e-06,
1174
+ "loss": 0.7971,
1175
+ "step": 1900
1176
+ },
1177
+ {
1178
+ "epoch": 3.21,
1179
+ "learning_rate": 9.53e-06,
1180
+ "loss": 0.7914,
1181
+ "step": 1910
1182
+ },
1183
+ {
1184
+ "epoch": 3.22,
1185
+ "learning_rate": 9.58e-06,
1186
+ "loss": 0.7873,
1187
+ "step": 1920
1188
+ },
1189
+ {
1190
+ "epoch": 3.24,
1191
+ "learning_rate": 9.630000000000001e-06,
1192
+ "loss": 0.7815,
1193
+ "step": 1930
1194
+ },
1195
+ {
1196
+ "epoch": 3.26,
1197
+ "learning_rate": 9.68e-06,
1198
+ "loss": 0.7958,
1199
+ "step": 1940
1200
+ },
1201
+ {
1202
+ "epoch": 3.27,
1203
+ "learning_rate": 9.73e-06,
1204
+ "loss": 0.7875,
1205
+ "step": 1950
1206
+ },
1207
+ {
1208
+ "epoch": 3.29,
1209
+ "learning_rate": 9.780000000000001e-06,
1210
+ "loss": 0.7968,
1211
+ "step": 1960
1212
+ },
1213
+ {
1214
+ "epoch": 3.31,
1215
+ "learning_rate": 9.83e-06,
1216
+ "loss": 0.7834,
1217
+ "step": 1970
1218
+ },
1219
+ {
1220
+ "epoch": 3.32,
1221
+ "learning_rate": 9.88e-06,
1222
+ "loss": 0.7935,
1223
+ "step": 1980
1224
+ },
1225
+ {
1226
+ "epoch": 3.34,
1227
+ "learning_rate": 9.930000000000001e-06,
1228
+ "loss": 0.7894,
1229
+ "step": 1990
1230
+ },
1231
+ {
1232
+ "epoch": 3.36,
1233
+ "learning_rate": 9.980000000000001e-06,
1234
+ "loss": 0.7908,
1235
+ "step": 2000
1236
+ },
1237
+ {
1238
+ "epoch": 3.36,
1239
+ "eval_code_evol_accuracy": 0.7468688781507118,
1240
+ "eval_code_evol_loss": 0.9775390625,
1241
+ "eval_code_evol_runtime": 44.2366,
1242
+ "eval_code_evol_samples_per_second": 17.429,
1243
+ "eval_code_evol_steps_per_second": 2.193,
1244
+ "step": 2000
1245
+ }
1246
+ ],
1247
+ "max_steps": 4760,
1248
+ "num_train_epochs": 8,
1249
+ "total_flos": 832756800815104.0,
1250
+ "trial_name": null,
1251
+ "trial_params": null
1252
+ }