Hugging Face's logo

winddude
/
wizardLM-LlaMA-LoRA-13

English
Model card Files Community
winddude commited on
Commit
acd2f6c
•
1 Parent(s): 66bdb77

init upload

Browse files
Files changed (35) hide show
  1. adapter_config.json +17 -0
  2. adapter_model.bin +3 -0
  3. checkpoint-1200/optimizer.pt +3 -0
  4. checkpoint-1200/pytorch_model.bin +3 -0
  5. checkpoint-1200/rng_state_0.pth +3 -0
  6. checkpoint-1200/rng_state_1.pth +3 -0
  7. checkpoint-1200/rng_state_2.pth +3 -0
  8. checkpoint-1200/rng_state_3.pth +3 -0
  9. checkpoint-1200/rng_state_4.pth +3 -0
  10. checkpoint-1200/scaler.pt +3 -0
  11. checkpoint-1200/scheduler.pt +3 -0
  12. checkpoint-1200/trainer_state.json +784 -0
  13. checkpoint-1200/training_args.bin +3 -0
  14. checkpoint-1400/optimizer.pt +3 -0
  15. checkpoint-1400/pytorch_model.bin +3 -0
  16. checkpoint-1400/rng_state_0.pth +3 -0
  17. checkpoint-1400/rng_state_1.pth +3 -0
  18. checkpoint-1400/rng_state_2.pth +3 -0
  19. checkpoint-1400/rng_state_3.pth +3 -0
  20. checkpoint-1400/rng_state_4.pth +3 -0
  21. checkpoint-1400/scaler.pt +3 -0
  22. checkpoint-1400/scheduler.pt +3 -0
  23. checkpoint-1400/trainer_state.json +912 -0
  24. checkpoint-1400/training_args.bin +3 -0
  25. checkpoint-1600/optimizer.pt +3 -0
  26. checkpoint-1600/pytorch_model.bin +3 -0
  27. checkpoint-1600/rng_state_0.pth +3 -0
  28. checkpoint-1600/rng_state_1.pth +3 -0
  29. checkpoint-1600/rng_state_2.pth +3 -0
  30. checkpoint-1600/rng_state_3.pth +3 -0
  31. checkpoint-1600/rng_state_4.pth +3 -0
  32. checkpoint-1600/scaler.pt +3 -0
  33. checkpoint-1600/scheduler.pt +3 -0
  34. checkpoint-1600/trainer_state.json +1040 -0
  35. checkpoint-1600/training_args.bin +3 -0
adapter_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "/home/llmadmin/models/my_llama_hf/llama_hf_13B",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "lora_alpha": 16,
8
+ "lora_dropout": 0.05,
9
+ "modules_to_save": null,
10
+ "peft_type": "LORA",
11
+ "r": 8,
12
+ "target_modules": [
13
+ "q_proj",
14
+ "v_proj"
15
+ ],
16
+ "task_type": "CAUSAL_LM"
17
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e2ef2724c42d6d72cda384f2c0abb699e5f7d5edd3a4d003cf69c838ca1a200
3
+ size 26271757
checkpoint-1200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d08ed72b7540396484932dd3e686b7596055f5fdd46bfc09fccc244e9b965651
3
+ size 52562757
checkpoint-1200/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8a547491487ac9a8a833ccc6a0905062a1b73c2446422109cdf4e71a3d45066
3
+ size 13370465717
checkpoint-1200/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2093c7c9715a1441d1ac27504d63e14d1ebec05e1f68890309474baa68c9774
3
+ size 14583
checkpoint-1200/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd5b22c84e297b181cca80136e185d28ee8e38f84faf8b18140530403516089a
3
+ size 14583
checkpoint-1200/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b638dcb44483ed7a9eea7ffbdd09fe93522f245e691e151265f49754a30da5b2
3
+ size 14583
checkpoint-1200/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:379b59420d4c1957570f2c4b40cf4b56b8020e88acaca2a53ec455c8247dd652
3
+ size 14583
checkpoint-1200/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cea30bb912a3009427f48f3bb697480351077f82f85256f0d2e42fdde577c920
3
+ size 14583
checkpoint-1200/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee008b0a1f8419254f7f60819effec8972bdac102c2d6c7468c6c898e2991119
3
+ size 557
checkpoint-1200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06a5a2c3d591f95bb74d8f5a26deb1fc6671ffa6e71b80bd16adcb97e82200b9
3
+ size 627
checkpoint-1200/trainer_state.json ADDED
@@ -0,0 +1,784 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6952534317970276,
3
+ "best_model_checkpoint": "/home/llmadmin/models/loras-trained/wizardLM-llama-lora-13b/checkpoint-1200",
4
+ "epoch": 2.1176470588235294,
5
+ "global_step": 1200,
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": 2.6999999999999996e-05,
13
+ "loss": 1.2009,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.04,
18
+ "learning_rate": 5.6999999999999996e-05,
19
+ "loss": 1.198,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.05,
24
+ "learning_rate": 8.699999999999999e-05,
25
+ "loss": 1.1318,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.07,
30
+ "learning_rate": 0.000117,
31
+ "loss": 1.0446,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.09,
36
+ "learning_rate": 0.000147,
37
+ "loss": 0.9254,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.11,
42
+ "learning_rate": 0.00017699999999999997,
43
+ "loss": 0.8364,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "learning_rate": 0.00020699999999999996,
49
+ "loss": 0.818,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.14,
54
+ "learning_rate": 0.000237,
55
+ "loss": 0.8094,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.16,
60
+ "learning_rate": 0.000267,
61
+ "loss": 0.7891,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.18,
66
+ "learning_rate": 0.00029699999999999996,
67
+ "loss": 0.7915,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.19,
72
+ "learning_rate": 0.0002983103879849812,
73
+ "loss": 0.777,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.21,
78
+ "learning_rate": 0.000296433041301627,
79
+ "loss": 0.7809,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.23,
84
+ "learning_rate": 0.0002945556946182728,
85
+ "loss": 0.7705,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.25,
90
+ "learning_rate": 0.0002926783479349186,
91
+ "loss": 0.7545,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.26,
96
+ "learning_rate": 0.00029080100125156443,
97
+ "loss": 0.7584,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.28,
102
+ "learning_rate": 0.00028892365456821025,
103
+ "loss": 0.751,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.3,
108
+ "learning_rate": 0.00028704630788485607,
109
+ "loss": 0.757,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.32,
114
+ "learning_rate": 0.00028516896120150183,
115
+ "loss": 0.7446,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.34,
120
+ "learning_rate": 0.00028329161451814765,
121
+ "loss": 0.7411,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.35,
126
+ "learning_rate": 0.0002814142678347935,
127
+ "loss": 0.754,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.35,
132
+ "eval_loss": 0.7339245676994324,
133
+ "eval_runtime": 65.2861,
134
+ "eval_samples_per_second": 30.634,
135
+ "eval_steps_per_second": 0.766,
136
+ "step": 200
137
+ },
138
+ {
139
+ "epoch": 0.37,
140
+ "learning_rate": 0.00027953692115143924,
141
+ "loss": 0.7487,
142
+ "step": 210
143
+ },
144
+ {
145
+ "epoch": 0.39,
146
+ "learning_rate": 0.00027765957446808506,
147
+ "loss": 0.7478,
148
+ "step": 220
149
+ },
150
+ {
151
+ "epoch": 0.41,
152
+ "learning_rate": 0.0002757822277847309,
153
+ "loss": 0.7336,
154
+ "step": 230
155
+ },
156
+ {
157
+ "epoch": 0.42,
158
+ "learning_rate": 0.0002739048811013767,
159
+ "loss": 0.7386,
160
+ "step": 240
161
+ },
162
+ {
163
+ "epoch": 0.44,
164
+ "learning_rate": 0.0002720275344180225,
165
+ "loss": 0.7359,
166
+ "step": 250
167
+ },
168
+ {
169
+ "epoch": 0.46,
170
+ "learning_rate": 0.0002701501877346683,
171
+ "loss": 0.7455,
172
+ "step": 260
173
+ },
174
+ {
175
+ "epoch": 0.48,
176
+ "learning_rate": 0.0002682728410513141,
177
+ "loss": 0.733,
178
+ "step": 270
179
+ },
180
+ {
181
+ "epoch": 0.49,
182
+ "learning_rate": 0.0002663954943679599,
183
+ "loss": 0.7332,
184
+ "step": 280
185
+ },
186
+ {
187
+ "epoch": 0.51,
188
+ "learning_rate": 0.00026451814768460575,
189
+ "loss": 0.7253,
190
+ "step": 290
191
+ },
192
+ {
193
+ "epoch": 0.53,
194
+ "learning_rate": 0.00026264080100125157,
195
+ "loss": 0.732,
196
+ "step": 300
197
+ },
198
+ {
199
+ "epoch": 0.55,
200
+ "learning_rate": 0.00026076345431789733,
201
+ "loss": 0.7271,
202
+ "step": 310
203
+ },
204
+ {
205
+ "epoch": 0.56,
206
+ "learning_rate": 0.00025888610763454315,
207
+ "loss": 0.7365,
208
+ "step": 320
209
+ },
210
+ {
211
+ "epoch": 0.58,
212
+ "learning_rate": 0.00025700876095118897,
213
+ "loss": 0.7266,
214
+ "step": 330
215
+ },
216
+ {
217
+ "epoch": 0.6,
218
+ "learning_rate": 0.0002551314142678348,
219
+ "loss": 0.733,
220
+ "step": 340
221
+ },
222
+ {
223
+ "epoch": 0.62,
224
+ "learning_rate": 0.00025325406758448056,
225
+ "loss": 0.7343,
226
+ "step": 350
227
+ },
228
+ {
229
+ "epoch": 0.64,
230
+ "learning_rate": 0.0002513767209011264,
231
+ "loss": 0.723,
232
+ "step": 360
233
+ },
234
+ {
235
+ "epoch": 0.65,
236
+ "learning_rate": 0.0002494993742177722,
237
+ "loss": 0.7391,
238
+ "step": 370
239
+ },
240
+ {
241
+ "epoch": 0.67,
242
+ "learning_rate": 0.000247622027534418,
243
+ "loss": 0.7265,
244
+ "step": 380
245
+ },
246
+ {
247
+ "epoch": 0.69,
248
+ "learning_rate": 0.00024574468085106384,
249
+ "loss": 0.7283,
250
+ "step": 390
251
+ },
252
+ {
253
+ "epoch": 0.71,
254
+ "learning_rate": 0.0002438673341677096,
255
+ "loss": 0.7306,
256
+ "step": 400
257
+ },
258
+ {
259
+ "epoch": 0.71,
260
+ "eval_loss": 0.7154878377914429,
261
+ "eval_runtime": 65.4382,
262
+ "eval_samples_per_second": 30.563,
263
+ "eval_steps_per_second": 0.764,
264
+ "step": 400
265
+ },
266
+ {
267
+ "epoch": 0.72,
268
+ "learning_rate": 0.00024198998748435542,
269
+ "loss": 0.725,
270
+ "step": 410
271
+ },
272
+ {
273
+ "epoch": 0.74,
274
+ "learning_rate": 0.00024011264080100122,
275
+ "loss": 0.7275,
276
+ "step": 420
277
+ },
278
+ {
279
+ "epoch": 0.76,
280
+ "learning_rate": 0.000238235294117647,
281
+ "loss": 0.7188,
282
+ "step": 430
283
+ },
284
+ {
285
+ "epoch": 0.78,
286
+ "learning_rate": 0.00023635794743429286,
287
+ "loss": 0.7216,
288
+ "step": 440
289
+ },
290
+ {
291
+ "epoch": 0.79,
292
+ "learning_rate": 0.00023448060075093865,
293
+ "loss": 0.7132,
294
+ "step": 450
295
+ },
296
+ {
297
+ "epoch": 0.81,
298
+ "learning_rate": 0.00023260325406758447,
299
+ "loss": 0.7372,
300
+ "step": 460
301
+ },
302
+ {
303
+ "epoch": 0.83,
304
+ "learning_rate": 0.00023072590738423026,
305
+ "loss": 0.7111,
306
+ "step": 470
307
+ },
308
+ {
309
+ "epoch": 0.85,
310
+ "learning_rate": 0.00022884856070087608,
311
+ "loss": 0.7131,
312
+ "step": 480
313
+ },
314
+ {
315
+ "epoch": 0.86,
316
+ "learning_rate": 0.00022697121401752188,
317
+ "loss": 0.7193,
318
+ "step": 490
319
+ },
320
+ {
321
+ "epoch": 0.88,
322
+ "learning_rate": 0.00022509386733416767,
323
+ "loss": 0.7152,
324
+ "step": 500
325
+ },
326
+ {
327
+ "epoch": 0.9,
328
+ "learning_rate": 0.00022321652065081352,
329
+ "loss": 0.7234,
330
+ "step": 510
331
+ },
332
+ {
333
+ "epoch": 0.92,
334
+ "learning_rate": 0.0002213391739674593,
335
+ "loss": 0.7193,
336
+ "step": 520
337
+ },
338
+ {
339
+ "epoch": 0.94,
340
+ "learning_rate": 0.00021946182728410513,
341
+ "loss": 0.7208,
342
+ "step": 530
343
+ },
344
+ {
345
+ "epoch": 0.95,
346
+ "learning_rate": 0.00021758448060075092,
347
+ "loss": 0.7205,
348
+ "step": 540
349
+ },
350
+ {
351
+ "epoch": 0.97,
352
+ "learning_rate": 0.00021570713391739672,
353
+ "loss": 0.7154,
354
+ "step": 550
355
+ },
356
+ {
357
+ "epoch": 0.99,
358
+ "learning_rate": 0.00021382978723404254,
359
+ "loss": 0.7224,
360
+ "step": 560
361
+ },
362
+ {
363
+ "epoch": 1.01,
364
+ "learning_rate": 0.00021195244055068833,
365
+ "loss": 0.7015,
366
+ "step": 570
367
+ },
368
+ {
369
+ "epoch": 1.02,
370
+ "learning_rate": 0.00021007509386733418,
371
+ "loss": 0.7168,
372
+ "step": 580
373
+ },
374
+ {
375
+ "epoch": 1.04,
376
+ "learning_rate": 0.00020819774718397997,
377
+ "loss": 0.7281,
378
+ "step": 590
379
+ },
380
+ {
381
+ "epoch": 1.06,
382
+ "learning_rate": 0.00020632040050062576,
383
+ "loss": 0.7112,
384
+ "step": 600
385
+ },
386
+ {
387
+ "epoch": 1.06,
388
+ "eval_loss": 0.7068994045257568,
389
+ "eval_runtime": 65.2314,
390
+ "eval_samples_per_second": 30.66,
391
+ "eval_steps_per_second": 0.767,
392
+ "step": 600
393
+ },
394
+ {
395
+ "epoch": 1.08,
396
+ "learning_rate": 0.00020444305381727158,
397
+ "loss": 0.7133,
398
+ "step": 610
399
+ },
400
+ {
401
+ "epoch": 1.09,
402
+ "learning_rate": 0.00020256570713391738,
403
+ "loss": 0.7047,
404
+ "step": 620
405
+ },
406
+ {
407
+ "epoch": 1.11,
408
+ "learning_rate": 0.00020068836045056317,
409
+ "loss": 0.7163,
410
+ "step": 630
411
+ },
412
+ {
413
+ "epoch": 1.13,
414
+ "learning_rate": 0.000198811013767209,
415
+ "loss": 0.7168,
416
+ "step": 640
417
+ },
418
+ {
419
+ "epoch": 1.15,
420
+ "learning_rate": 0.00019693366708385478,
421
+ "loss": 0.7028,
422
+ "step": 650
423
+ },
424
+ {
425
+ "epoch": 1.16,
426
+ "learning_rate": 0.00019505632040050063,
427
+ "loss": 0.7064,
428
+ "step": 660
429
+ },
430
+ {
431
+ "epoch": 1.18,
432
+ "learning_rate": 0.00019317897371714642,
433
+ "loss": 0.7127,
434
+ "step": 670
435
+ },
436
+ {
437
+ "epoch": 1.2,
438
+ "learning_rate": 0.00019130162703379222,
439
+ "loss": 0.7193,
440
+ "step": 680
441
+ },
442
+ {
443
+ "epoch": 1.22,
444
+ "learning_rate": 0.00018942428035043804,
445
+ "loss": 0.6966,
446
+ "step": 690
447
+ },
448
+ {
449
+ "epoch": 1.24,
450
+ "learning_rate": 0.00018754693366708383,
451
+ "loss": 0.7104,
452
+ "step": 700
453
+ },
454
+ {
455
+ "epoch": 1.25,
456
+ "learning_rate": 0.00018566958698372962,
457
+ "loss": 0.7076,
458
+ "step": 710
459
+ },
460
+ {
461
+ "epoch": 1.27,
462
+ "learning_rate": 0.00018379224030037544,
463
+ "loss": 0.6971,
464
+ "step": 720
465
+ },
466
+ {
467
+ "epoch": 1.29,
468
+ "learning_rate": 0.00018191489361702126,
469
+ "loss": 0.6941,
470
+ "step": 730
471
+ },
472
+ {
473
+ "epoch": 1.31,
474
+ "learning_rate": 0.00018003754693366708,
475
+ "loss": 0.6992,
476
+ "step": 740
477
+ },
478
+ {
479
+ "epoch": 1.32,
480
+ "learning_rate": 0.00017816020025031287,
481
+ "loss": 0.7059,
482
+ "step": 750
483
+ },
484
+ {
485
+ "epoch": 1.34,
486
+ "learning_rate": 0.0001762828535669587,
487
+ "loss": 0.6936,
488
+ "step": 760
489
+ },
490
+ {
491
+ "epoch": 1.36,
492
+ "learning_rate": 0.0001744055068836045,
493
+ "loss": 0.6961,
494
+ "step": 770
495
+ },
496
+ {
497
+ "epoch": 1.38,
498
+ "learning_rate": 0.00017252816020025028,
499
+ "loss": 0.7125,
500
+ "step": 780
501
+ },
502
+ {
503
+ "epoch": 1.39,
504
+ "learning_rate": 0.0001706508135168961,
505
+ "loss": 0.7046,
506
+ "step": 790
507
+ },
508
+ {
509
+ "epoch": 1.41,
510
+ "learning_rate": 0.00016877346683354192,
511
+ "loss": 0.7034,
512
+ "step": 800
513
+ },
514
+ {
515
+ "epoch": 1.41,
516
+ "eval_loss": 0.7016428709030151,
517
+ "eval_runtime": 65.2245,
518
+ "eval_samples_per_second": 30.663,
519
+ "eval_steps_per_second": 0.767,
520
+ "step": 800
521
+ },
522
+ {
523
+ "epoch": 1.43,
524
+ "learning_rate": 0.00016689612015018774,
525
+ "loss": 0.7011,
526
+ "step": 810
527
+ },
528
+ {
529
+ "epoch": 1.45,
530
+ "learning_rate": 0.00016501877346683353,
531
+ "loss": 0.7049,
532
+ "step": 820
533
+ },
534
+ {
535
+ "epoch": 1.46,
536
+ "learning_rate": 0.00016314142678347933,
537
+ "loss": 0.6965,
538
+ "step": 830
539
+ },
540
+ {
541
+ "epoch": 1.48,
542
+ "learning_rate": 0.00016126408010012515,
543
+ "loss": 0.6965,
544
+ "step": 840
545
+ },
546
+ {
547
+ "epoch": 1.5,
548
+ "learning_rate": 0.00015938673341677094,
549
+ "loss": 0.6975,
550
+ "step": 850
551
+ },
552
+ {
553
+ "epoch": 1.52,
554
+ "learning_rate": 0.00015750938673341673,
555
+ "loss": 0.7094,
556
+ "step": 860
557
+ },
558
+ {
559
+ "epoch": 1.54,
560
+ "learning_rate": 0.00015563204005006258,
561
+ "loss": 0.6918,
562
+ "step": 870
563
+ },
564
+ {
565
+ "epoch": 1.55,
566
+ "learning_rate": 0.00015375469336670837,
567
+ "loss": 0.7031,
568
+ "step": 880
569
+ },
570
+ {
571
+ "epoch": 1.57,
572
+ "learning_rate": 0.0001518773466833542,
573
+ "loss": 0.7027,
574
+ "step": 890
575
+ },
576
+ {
577
+ "epoch": 1.59,
578
+ "learning_rate": 0.00015,
579
+ "loss": 0.6996,
580
+ "step": 900
581
+ },
582
+ {
583
+ "epoch": 1.61,
584
+ "learning_rate": 0.00014812265331664578,
585
+ "loss": 0.7048,
586
+ "step": 910
587
+ },
588
+ {
589
+ "epoch": 1.62,
590
+ "learning_rate": 0.0001462453066332916,
591
+ "loss": 0.6962,
592
+ "step": 920
593
+ },
594
+ {
595
+ "epoch": 1.64,
596
+ "learning_rate": 0.00014436795994993742,
597
+ "loss": 0.6918,
598
+ "step": 930
599
+ },
600
+ {
601
+ "epoch": 1.66,
602
+ "learning_rate": 0.0001424906132665832,
603
+ "loss": 0.6979,
604
+ "step": 940
605
+ },
606
+ {
607
+ "epoch": 1.68,
608
+ "learning_rate": 0.000140613266583229,
609
+ "loss": 0.6958,
610
+ "step": 950
611
+ },
612
+ {
613
+ "epoch": 1.69,
614
+ "learning_rate": 0.00013873591989987483,
615
+ "loss": 0.7027,
616
+ "step": 960
617
+ },
618
+ {
619
+ "epoch": 1.71,
620
+ "learning_rate": 0.00013685857321652065,
621
+ "loss": 0.7092,
622
+ "step": 970
623
+ },
624
+ {
625
+ "epoch": 1.73,
626
+ "learning_rate": 0.00013498122653316644,
627
+ "loss": 0.6917,
628
+ "step": 980
629
+ },
630
+ {
631
+ "epoch": 1.75,
632
+ "learning_rate": 0.00013310387984981226,
633
+ "loss": 0.6996,
634
+ "step": 990
635
+ },
636
+ {
637
+ "epoch": 1.76,
638
+ "learning_rate": 0.00013122653316645805,
639
+ "loss": 0.6889,
640
+ "step": 1000
641
+ },
642
+ {
643
+ "epoch": 1.76,
644
+ "eval_loss": 0.6975076794624329,
645
+ "eval_runtime": 65.2345,
646
+ "eval_samples_per_second": 30.659,
647
+ "eval_steps_per_second": 0.766,
648
+ "step": 1000
649
+ },
650
+ {
651
+ "epoch": 1.78,
652
+ "learning_rate": 0.00012934918648310387,
653
+ "loss": 0.7083,
654
+ "step": 1010
655
+ },
656
+ {
657
+ "epoch": 1.8,
658
+ "learning_rate": 0.00012747183979974967,
659
+ "loss": 0.6934,
660
+ "step": 1020
661
+ },
662
+ {
663
+ "epoch": 1.82,
664
+ "learning_rate": 0.00012559449311639549,
665
+ "loss": 0.7002,
666
+ "step": 1030
667
+ },
668
+ {
669
+ "epoch": 1.84,
670
+ "learning_rate": 0.00012371714643304128,
671
+ "loss": 0.6899,
672
+ "step": 1040
673
+ },
674
+ {
675
+ "epoch": 1.85,
676
+ "learning_rate": 0.0001218397997496871,
677
+ "loss": 0.6856,
678
+ "step": 1050
679
+ },
680
+ {
681
+ "epoch": 1.87,
682
+ "learning_rate": 0.00011996245306633289,
683
+ "loss": 0.695,
684
+ "step": 1060
685
+ },
686
+ {
687
+ "epoch": 1.89,
688
+ "learning_rate": 0.00011808510638297871,
689
+ "loss": 0.6934,
690
+ "step": 1070
691
+ },
692
+ {
693
+ "epoch": 1.91,
694
+ "learning_rate": 0.00011620775969962452,
695
+ "loss": 0.6933,
696
+ "step": 1080
697
+ },
698
+ {
699
+ "epoch": 1.92,
700
+ "learning_rate": 0.00011433041301627033,
701
+ "loss": 0.6983,
702
+ "step": 1090
703
+ },
704
+ {
705
+ "epoch": 1.94,
706
+ "learning_rate": 0.00011245306633291615,
707
+ "loss": 0.705,
708
+ "step": 1100
709
+ },
710
+ {
711
+ "epoch": 1.96,
712
+ "learning_rate": 0.00011057571964956194,
713
+ "loss": 0.6944,
714
+ "step": 1110
715
+ },
716
+ {
717
+ "epoch": 1.98,
718
+ "learning_rate": 0.00010869837296620774,
719
+ "loss": 0.6977,
720
+ "step": 1120
721
+ },
722
+ {
723
+ "epoch": 1.99,
724
+ "learning_rate": 0.00010682102628285355,
725
+ "loss": 0.6966,
726
+ "step": 1130
727
+ },
728
+ {
729
+ "epoch": 2.01,
730
+ "learning_rate": 0.00010494367959949937,
731
+ "loss": 0.6868,
732
+ "step": 1140
733
+ },
734
+ {
735
+ "epoch": 2.03,
736
+ "learning_rate": 0.00010306633291614518,
737
+ "loss": 0.6802,
738
+ "step": 1150
739
+ },
740
+ {
741
+ "epoch": 2.05,
742
+ "learning_rate": 0.00010118898623279097,
743
+ "loss": 0.6916,
744
+ "step": 1160
745
+ },
746
+ {
747
+ "epoch": 2.06,
748
+ "learning_rate": 9.931163954943679e-05,
749
+ "loss": 0.7018,
750
+ "step": 1170
751
+ },
752
+ {
753
+ "epoch": 2.08,
754
+ "learning_rate": 9.74342928660826e-05,
755
+ "loss": 0.6951,
756
+ "step": 1180
757
+ },
758
+ {
759
+ "epoch": 2.1,
760
+ "learning_rate": 9.55569461827284e-05,
761
+ "loss": 0.6906,
762
+ "step": 1190
763
+ },
764
+ {
765
+ "epoch": 2.12,
766
+ "learning_rate": 9.36795994993742e-05,
767
+ "loss": 0.6965,
768
+ "step": 1200
769
+ },
770
+ {
771
+ "epoch": 2.12,
772
+ "eval_loss": 0.6952534317970276,
773
+ "eval_runtime": 65.233,
774
+ "eval_samples_per_second": 30.659,
775
+ "eval_steps_per_second": 0.766,
776
+ "step": 1200
777
+ }
778
+ ],
779
+ "max_steps": 1698,
780
+ "num_train_epochs": 3,
781
+ "total_flos": 5.549133471969968e+18,
782
+ "trial_name": null,
783
+ "trial_params": null
784
+ }
checkpoint-1200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3475910ba01356354994b8c54e7272fa9f9ab35fefffcb1f6084a7aa7bd2ffae
3
+ size 3643
checkpoint-1400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ec52c77ec209e9a55a502a151a915e7acad360fc13d9da3afc40ded25abdb75
3
+ size 52562757
checkpoint-1400/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ba198dba98af7761ff5e909f4a9d2a76c471799780dce13d25adb6f8bdb0287
3
+ size 13370465717
checkpoint-1400/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d29b7d2be2c9c8a1169cce377e35172ee8ee5227486a5ce859f3bf885325c86a
3
+ size 14583
checkpoint-1400/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45570a0f85ceb33c6d0a6675da2b8841531d6424b7a1a14fd98378c6e79ea537
3
+ size 14583
checkpoint-1400/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:015f85483697dc1afc6aa584668d7ff58935e956f53f9af5e83c4afa574423dd
3
+ size 14583
checkpoint-1400/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d24ca681e50e2e2bdd4ab507a06e2a433d178bf8d565a967c94ba89354b8ea03
3
+ size 14583
checkpoint-1400/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:918da24e5e59bb3417a806298d3cdcd617ec779405de2542ed5f0df78cfe0dd5
3
+ size 14583
checkpoint-1400/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a82da4d58c0a642b7a026643bf2fd0532d55449337f268e519bc6f62f002427e
3
+ size 557
checkpoint-1400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e821e3f7f8f98071668a10bad849c2a467d65aec6a2a82198d02a280a00cdd47
3
+ size 627
checkpoint-1400/trainer_state.json ADDED
@@ -0,0 +1,912 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6928293704986572,
3
+ "best_model_checkpoint": "/home/llmadmin/models/loras-trained/wizardLM-llama-lora-13b/checkpoint-1400",
4
+ "epoch": 2.4705882352941178,
5
+ "global_step": 1400,
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": 2.6999999999999996e-05,
13
+ "loss": 1.2009,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.04,
18
+ "learning_rate": 5.6999999999999996e-05,
19
+ "loss": 1.198,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.05,
24
+ "learning_rate": 8.699999999999999e-05,
25
+ "loss": 1.1318,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.07,
30
+ "learning_rate": 0.000117,
31
+ "loss": 1.0446,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.09,
36
+ "learning_rate": 0.000147,
37
+ "loss": 0.9254,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.11,
42
+ "learning_rate": 0.00017699999999999997,
43
+ "loss": 0.8364,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "learning_rate": 0.00020699999999999996,
49
+ "loss": 0.818,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.14,
54
+ "learning_rate": 0.000237,
55
+ "loss": 0.8094,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.16,
60
+ "learning_rate": 0.000267,
61
+ "loss": 0.7891,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.18,
66
+ "learning_rate": 0.00029699999999999996,
67
+ "loss": 0.7915,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.19,
72
+ "learning_rate": 0.0002983103879849812,
73
+ "loss": 0.777,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.21,
78
+ "learning_rate": 0.000296433041301627,
79
+ "loss": 0.7809,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.23,
84
+ "learning_rate": 0.0002945556946182728,
85
+ "loss": 0.7705,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.25,
90
+ "learning_rate": 0.0002926783479349186,
91
+ "loss": 0.7545,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.26,
96
+ "learning_rate": 0.00029080100125156443,
97
+ "loss": 0.7584,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.28,
102
+ "learning_rate": 0.00028892365456821025,
103
+ "loss": 0.751,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.3,
108
+ "learning_rate": 0.00028704630788485607,
109
+ "loss": 0.757,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.32,
114
+ "learning_rate": 0.00028516896120150183,
115
+ "loss": 0.7446,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.34,
120
+ "learning_rate": 0.00028329161451814765,
121
+ "loss": 0.7411,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.35,
126
+ "learning_rate": 0.0002814142678347935,
127
+ "loss": 0.754,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.35,
132
+ "eval_loss": 0.7339245676994324,
133
+ "eval_runtime": 65.2861,
134
+ "eval_samples_per_second": 30.634,
135
+ "eval_steps_per_second": 0.766,
136
+ "step": 200
137
+ },
138
+ {
139
+ "epoch": 0.37,
140
+ "learning_rate": 0.00027953692115143924,
141
+ "loss": 0.7487,
142
+ "step": 210
143
+ },
144
+ {
145
+ "epoch": 0.39,
146
+ "learning_rate": 0.00027765957446808506,
147
+ "loss": 0.7478,
148
+ "step": 220
149
+ },
150
+ {
151
+ "epoch": 0.41,
152
+ "learning_rate": 0.0002757822277847309,
153
+ "loss": 0.7336,
154
+ "step": 230
155
+ },
156
+ {
157
+ "epoch": 0.42,
158
+ "learning_rate": 0.0002739048811013767,
159
+ "loss": 0.7386,
160
+ "step": 240
161
+ },
162
+ {
163
+ "epoch": 0.44,
164
+ "learning_rate": 0.0002720275344180225,
165
+ "loss": 0.7359,
166
+ "step": 250
167
+ },
168
+ {
169
+ "epoch": 0.46,
170
+ "learning_rate": 0.0002701501877346683,
171
+ "loss": 0.7455,
172
+ "step": 260
173
+ },
174
+ {
175
+ "epoch": 0.48,
176
+ "learning_rate": 0.0002682728410513141,
177
+ "loss": 0.733,
178
+ "step": 270
179
+ },
180
+ {
181
+ "epoch": 0.49,
182
+ "learning_rate": 0.0002663954943679599,
183
+ "loss": 0.7332,
184
+ "step": 280
185
+ },
186
+ {
187
+ "epoch": 0.51,
188
+ "learning_rate": 0.00026451814768460575,
189
+ "loss": 0.7253,
190
+ "step": 290
191
+ },
192
+ {
193
+ "epoch": 0.53,
194
+ "learning_rate": 0.00026264080100125157,
195
+ "loss": 0.732,
196
+ "step": 300
197
+ },
198
+ {
199
+ "epoch": 0.55,
200
+ "learning_rate": 0.00026076345431789733,
201
+ "loss": 0.7271,
202
+ "step": 310
203
+ },
204
+ {
205
+ "epoch": 0.56,
206
+ "learning_rate": 0.00025888610763454315,
207
+ "loss": 0.7365,
208
+ "step": 320
209
+ },
210
+ {
211
+ "epoch": 0.58,
212
+ "learning_rate": 0.00025700876095118897,
213
+ "loss": 0.7266,
214
+ "step": 330
215
+ },
216
+ {
217
+ "epoch": 0.6,
218
+ "learning_rate": 0.0002551314142678348,
219
+ "loss": 0.733,
220
+ "step": 340
221
+ },
222
+ {
223
+ "epoch": 0.62,
224
+ "learning_rate": 0.00025325406758448056,
225
+ "loss": 0.7343,
226
+ "step": 350
227
+ },
228
+ {
229
+ "epoch": 0.64,
230
+ "learning_rate": 0.0002513767209011264,
231
+ "loss": 0.723,
232
+ "step": 360
233
+ },
234
+ {
235
+ "epoch": 0.65,
236
+ "learning_rate": 0.0002494993742177722,
237
+ "loss": 0.7391,
238
+ "step": 370
239
+ },
240
+ {
241
+ "epoch": 0.67,
242
+ "learning_rate": 0.000247622027534418,
243
+ "loss": 0.7265,
244
+ "step": 380
245
+ },
246
+ {
247
+ "epoch": 0.69,
248
+ "learning_rate": 0.00024574468085106384,
249
+ "loss": 0.7283,
250
+ "step": 390
251
+ },
252
+ {
253
+ "epoch": 0.71,
254
+ "learning_rate": 0.0002438673341677096,
255
+ "loss": 0.7306,
256
+ "step": 400
257
+ },
258
+ {
259
+ "epoch": 0.71,
260
+ "eval_loss": 0.7154878377914429,
261
+ "eval_runtime": 65.4382,
262
+ "eval_samples_per_second": 30.563,
263
+ "eval_steps_per_second": 0.764,
264
+ "step": 400
265
+ },
266
+ {
267
+ "epoch": 0.72,
268
+ "learning_rate": 0.00024198998748435542,
269
+ "loss": 0.725,
270
+ "step": 410
271
+ },
272
+ {
273
+ "epoch": 0.74,
274
+ "learning_rate": 0.00024011264080100122,
275
+ "loss": 0.7275,
276
+ "step": 420
277
+ },
278
+ {
279
+ "epoch": 0.76,
280
+ "learning_rate": 0.000238235294117647,
281
+ "loss": 0.7188,
282
+ "step": 430
283
+ },
284
+ {
285
+ "epoch": 0.78,
286
+ "learning_rate": 0.00023635794743429286,
287
+ "loss": 0.7216,
288
+ "step": 440
289
+ },
290
+ {
291
+ "epoch": 0.79,
292
+ "learning_rate": 0.00023448060075093865,
293
+ "loss": 0.7132,
294
+ "step": 450
295
+ },
296
+ {
297
+ "epoch": 0.81,
298
+ "learning_rate": 0.00023260325406758447,
299
+ "loss": 0.7372,
300
+ "step": 460
301
+ },
302
+ {
303
+ "epoch": 0.83,
304
+ "learning_rate": 0.00023072590738423026,
305
+ "loss": 0.7111,
306
+ "step": 470
307
+ },
308
+ {
309
+ "epoch": 0.85,
310
+ "learning_rate": 0.00022884856070087608,
311
+ "loss": 0.7131,
312
+ "step": 480
313
+ },
314
+ {
315
+ "epoch": 0.86,
316
+ "learning_rate": 0.00022697121401752188,
317
+ "loss": 0.7193,
318
+ "step": 490
319
+ },
320
+ {
321
+ "epoch": 0.88,
322
+ "learning_rate": 0.00022509386733416767,
323
+ "loss": 0.7152,
324
+ "step": 500
325
+ },
326
+ {
327
+ "epoch": 0.9,
328
+ "learning_rate": 0.00022321652065081352,
329
+ "loss": 0.7234,
330
+ "step": 510
331
+ },
332
+ {
333
+ "epoch": 0.92,
334
+ "learning_rate": 0.0002213391739674593,
335
+ "loss": 0.7193,
336
+ "step": 520
337
+ },
338
+ {
339
+ "epoch": 0.94,
340
+ "learning_rate": 0.00021946182728410513,
341
+ "loss": 0.7208,
342
+ "step": 530
343
+ },
344
+ {
345
+ "epoch": 0.95,
346
+ "learning_rate": 0.00021758448060075092,
347
+ "loss": 0.7205,
348
+ "step": 540
349
+ },
350
+ {
351
+ "epoch": 0.97,
352
+ "learning_rate": 0.00021570713391739672,
353
+ "loss": 0.7154,
354
+ "step": 550
355
+ },
356
+ {
357
+ "epoch": 0.99,
358
+ "learning_rate": 0.00021382978723404254,
359
+ "loss": 0.7224,
360
+ "step": 560
361
+ },
362
+ {
363
+ "epoch": 1.01,
364
+ "learning_rate": 0.00021195244055068833,
365
+ "loss": 0.7015,
366
+ "step": 570
367
+ },
368
+ {
369
+ "epoch": 1.02,
370
+ "learning_rate": 0.00021007509386733418,
371
+ "loss": 0.7168,
372
+ "step": 580
373
+ },
374
+ {
375
+ "epoch": 1.04,
376
+ "learning_rate": 0.00020819774718397997,
377
+ "loss": 0.7281,
378
+ "step": 590
379
+ },
380
+ {
381
+ "epoch": 1.06,
382
+ "learning_rate": 0.00020632040050062576,
383
+ "loss": 0.7112,
384
+ "step": 600
385
+ },
386
+ {
387
+ "epoch": 1.06,
388
+ "eval_loss": 0.7068994045257568,
389
+ "eval_runtime": 65.2314,
390
+ "eval_samples_per_second": 30.66,
391
+ "eval_steps_per_second": 0.767,
392
+ "step": 600
393
+ },
394
+ {
395
+ "epoch": 1.08,
396
+ "learning_rate": 0.00020444305381727158,
397
+ "loss": 0.7133,
398
+ "step": 610
399
+ },
400
+ {
401
+ "epoch": 1.09,
402
+ "learning_rate": 0.00020256570713391738,
403
+ "loss": 0.7047,
404
+ "step": 620
405
+ },
406
+ {
407
+ "epoch": 1.11,
408
+ "learning_rate": 0.00020068836045056317,
409
+ "loss": 0.7163,
410
+ "step": 630
411
+ },
412
+ {
413
+ "epoch": 1.13,
414
+ "learning_rate": 0.000198811013767209,
415
+ "loss": 0.7168,
416
+ "step": 640
417
+ },
418
+ {
419
+ "epoch": 1.15,
420
+ "learning_rate": 0.00019693366708385478,
421
+ "loss": 0.7028,
422
+ "step": 650
423
+ },
424
+ {
425
+ "epoch": 1.16,
426
+ "learning_rate": 0.00019505632040050063,
427
+ "loss": 0.7064,
428
+ "step": 660
429
+ },
430
+ {
431
+ "epoch": 1.18,
432
+ "learning_rate": 0.00019317897371714642,
433
+ "loss": 0.7127,
434
+ "step": 670
435
+ },
436
+ {
437
+ "epoch": 1.2,
438
+ "learning_rate": 0.00019130162703379222,
439
+ "loss": 0.7193,
440
+ "step": 680
441
+ },
442
+ {
443
+ "epoch": 1.22,
444
+ "learning_rate": 0.00018942428035043804,
445
+ "loss": 0.6966,
446
+ "step": 690
447
+ },
448
+ {
449
+ "epoch": 1.24,
450
+ "learning_rate": 0.00018754693366708383,
451
+ "loss": 0.7104,
452
+ "step": 700
453
+ },
454
+ {
455
+ "epoch": 1.25,
456
+ "learning_rate": 0.00018566958698372962,
457
+ "loss": 0.7076,
458
+ "step": 710
459
+ },
460
+ {
461
+ "epoch": 1.27,
462
+ "learning_rate": 0.00018379224030037544,
463
+ "loss": 0.6971,
464
+ "step": 720
465
+ },
466
+ {
467
+ "epoch": 1.29,
468
+ "learning_rate": 0.00018191489361702126,
469
+ "loss": 0.6941,
470
+ "step": 730
471
+ },
472
+ {
473
+ "epoch": 1.31,
474
+ "learning_rate": 0.00018003754693366708,
475
+ "loss": 0.6992,
476
+ "step": 740
477
+ },
478
+ {
479
+ "epoch": 1.32,
480
+ "learning_rate": 0.00017816020025031287,
481
+ "loss": 0.7059,
482
+ "step": 750
483
+ },
484
+ {
485
+ "epoch": 1.34,
486
+ "learning_rate": 0.0001762828535669587,
487
+ "loss": 0.6936,
488
+ "step": 760
489
+ },
490
+ {
491
+ "epoch": 1.36,
492
+ "learning_rate": 0.0001744055068836045,
493
+ "loss": 0.6961,
494
+ "step": 770
495
+ },
496
+ {
497
+ "epoch": 1.38,
498
+ "learning_rate": 0.00017252816020025028,
499
+ "loss": 0.7125,
500
+ "step": 780
501
+ },
502
+ {
503
+ "epoch": 1.39,
504
+ "learning_rate": 0.0001706508135168961,
505
+ "loss": 0.7046,
506
+ "step": 790
507
+ },
508
+ {
509
+ "epoch": 1.41,
510
+ "learning_rate": 0.00016877346683354192,
511
+ "loss": 0.7034,
512
+ "step": 800
513
+ },
514
+ {
515
+ "epoch": 1.41,
516
+ "eval_loss": 0.7016428709030151,
517
+ "eval_runtime": 65.2245,
518
+ "eval_samples_per_second": 30.663,
519
+ "eval_steps_per_second": 0.767,
520
+ "step": 800
521
+ },
522
+ {
523
+ "epoch": 1.43,
524
+ "learning_rate": 0.00016689612015018774,
525
+ "loss": 0.7011,
526
+ "step": 810
527
+ },
528
+ {
529
+ "epoch": 1.45,
530
+ "learning_rate": 0.00016501877346683353,
531
+ "loss": 0.7049,
532
+ "step": 820
533
+ },
534
+ {
535
+ "epoch": 1.46,
536
+ "learning_rate": 0.00016314142678347933,
537
+ "loss": 0.6965,
538
+ "step": 830
539
+ },
540
+ {
541
+ "epoch": 1.48,
542
+ "learning_rate": 0.00016126408010012515,
543
+ "loss": 0.6965,
544
+ "step": 840
545
+ },
546
+ {
547
+ "epoch": 1.5,
548
+ "learning_rate": 0.00015938673341677094,
549
+ "loss": 0.6975,
550
+ "step": 850
551
+ },
552
+ {
553
+ "epoch": 1.52,
554
+ "learning_rate": 0.00015750938673341673,
555
+ "loss": 0.7094,
556
+ "step": 860
557
+ },
558
+ {
559
+ "epoch": 1.54,
560
+ "learning_rate": 0.00015563204005006258,
561
+ "loss": 0.6918,
562
+ "step": 870
563
+ },
564
+ {
565
+ "epoch": 1.55,
566
+ "learning_rate": 0.00015375469336670837,
567
+ "loss": 0.7031,
568
+ "step": 880
569
+ },
570
+ {
571
+ "epoch": 1.57,
572
+ "learning_rate": 0.0001518773466833542,
573
+ "loss": 0.7027,
574
+ "step": 890
575
+ },
576
+ {
577
+ "epoch": 1.59,
578
+ "learning_rate": 0.00015,
579
+ "loss": 0.6996,
580
+ "step": 900
581
+ },
582
+ {
583
+ "epoch": 1.61,
584
+ "learning_rate": 0.00014812265331664578,
585
+ "loss": 0.7048,
586
+ "step": 910
587
+ },
588
+ {
589
+ "epoch": 1.62,
590
+ "learning_rate": 0.0001462453066332916,
591
+ "loss": 0.6962,
592
+ "step": 920
593
+ },
594
+ {
595
+ "epoch": 1.64,
596
+ "learning_rate": 0.00014436795994993742,
597
+ "loss": 0.6918,
598
+ "step": 930
599
+ },
600
+ {
601
+ "epoch": 1.66,
602
+ "learning_rate": 0.0001424906132665832,
603
+ "loss": 0.6979,
604
+ "step": 940
605
+ },
606
+ {
607
+ "epoch": 1.68,
608
+ "learning_rate": 0.000140613266583229,
609
+ "loss": 0.6958,
610
+ "step": 950
611
+ },
612
+ {
613
+ "epoch": 1.69,
614
+ "learning_rate": 0.00013873591989987483,
615
+ "loss": 0.7027,
616
+ "step": 960
617
+ },
618
+ {
619
+ "epoch": 1.71,
620
+ "learning_rate": 0.00013685857321652065,
621
+ "loss": 0.7092,
622
+ "step": 970
623
+ },
624
+ {
625
+ "epoch": 1.73,
626
+ "learning_rate": 0.00013498122653316644,
627
+ "loss": 0.6917,
628
+ "step": 980
629
+ },
630
+ {
631
+ "epoch": 1.75,
632
+ "learning_rate": 0.00013310387984981226,
633
+ "loss": 0.6996,
634
+ "step": 990
635
+ },
636
+ {
637
+ "epoch": 1.76,
638
+ "learning_rate": 0.00013122653316645805,
639
+ "loss": 0.6889,
640
+ "step": 1000
641
+ },
642
+ {
643
+ "epoch": 1.76,
644
+ "eval_loss": 0.6975076794624329,
645
+ "eval_runtime": 65.2345,
646
+ "eval_samples_per_second": 30.659,
647
+ "eval_steps_per_second": 0.766,
648
+ "step": 1000
649
+ },
650
+ {
651
+ "epoch": 1.78,
652
+ "learning_rate": 0.00012934918648310387,
653
+ "loss": 0.7083,
654
+ "step": 1010
655
+ },
656
+ {
657
+ "epoch": 1.8,
658
+ "learning_rate": 0.00012747183979974967,
659
+ "loss": 0.6934,
660
+ "step": 1020
661
+ },
662
+ {
663
+ "epoch": 1.82,
664
+ "learning_rate": 0.00012559449311639549,
665
+ "loss": 0.7002,
666
+ "step": 1030
667
+ },
668
+ {
669
+ "epoch": 1.84,
670
+ "learning_rate": 0.00012371714643304128,
671
+ "loss": 0.6899,
672
+ "step": 1040
673
+ },
674
+ {
675
+ "epoch": 1.85,
676
+ "learning_rate": 0.0001218397997496871,
677
+ "loss": 0.6856,
678
+ "step": 1050
679
+ },
680
+ {
681
+ "epoch": 1.87,
682
+ "learning_rate": 0.00011996245306633289,
683
+ "loss": 0.695,
684
+ "step": 1060
685
+ },
686
+ {
687
+ "epoch": 1.89,
688
+ "learning_rate": 0.00011808510638297871,
689
+ "loss": 0.6934,
690
+ "step": 1070
691
+ },
692
+ {
693
+ "epoch": 1.91,
694
+ "learning_rate": 0.00011620775969962452,
695
+ "loss": 0.6933,
696
+ "step": 1080
697
+ },
698
+ {
699
+ "epoch": 1.92,
700
+ "learning_rate": 0.00011433041301627033,
701
+ "loss": 0.6983,
702
+ "step": 1090
703
+ },
704
+ {
705
+ "epoch": 1.94,
706
+ "learning_rate": 0.00011245306633291615,
707
+ "loss": 0.705,
708
+ "step": 1100
709
+ },
710
+ {
711
+ "epoch": 1.96,
712
+ "learning_rate": 0.00011057571964956194,
713
+ "loss": 0.6944,
714
+ "step": 1110
715
+ },
716
+ {
717
+ "epoch": 1.98,
718
+ "learning_rate": 0.00010869837296620774,
719
+ "loss": 0.6977,
720
+ "step": 1120
721
+ },
722
+ {
723
+ "epoch": 1.99,
724
+ "learning_rate": 0.00010682102628285355,
725
+ "loss": 0.6966,
726
+ "step": 1130
727
+ },
728
+ {
729
+ "epoch": 2.01,
730
+ "learning_rate": 0.00010494367959949937,
731
+ "loss": 0.6868,
732
+ "step": 1140
733
+ },
734
+ {
735
+ "epoch": 2.03,
736
+ "learning_rate": 0.00010306633291614518,
737
+ "loss": 0.6802,
738
+ "step": 1150
739
+ },
740
+ {
741
+ "epoch": 2.05,
742
+ "learning_rate": 0.00010118898623279097,
743
+ "loss": 0.6916,
744
+ "step": 1160
745
+ },
746
+ {
747
+ "epoch": 2.06,
748
+ "learning_rate": 9.931163954943679e-05,
749
+ "loss": 0.7018,
750
+ "step": 1170
751
+ },
752
+ {
753
+ "epoch": 2.08,
754
+ "learning_rate": 9.74342928660826e-05,
755
+ "loss": 0.6951,
756
+ "step": 1180
757
+ },
758
+ {
759
+ "epoch": 2.1,
760
+ "learning_rate": 9.55569461827284e-05,
761
+ "loss": 0.6906,
762
+ "step": 1190
763
+ },
764
+ {
765
+ "epoch": 2.12,
766
+ "learning_rate": 9.36795994993742e-05,
767
+ "loss": 0.6965,
768
+ "step": 1200
769
+ },
770
+ {
771
+ "epoch": 2.12,
772
+ "eval_loss": 0.6952534317970276,
773
+ "eval_runtime": 65.233,
774
+ "eval_samples_per_second": 30.659,
775
+ "eval_steps_per_second": 0.766,
776
+ "step": 1200
777
+ },
778
+ {
779
+ "epoch": 2.14,
780
+ "learning_rate": 9.180225281602002e-05,
781
+ "loss": 0.6969,
782
+ "step": 1210
783
+ },
784
+ {
785
+ "epoch": 2.15,
786
+ "learning_rate": 8.992490613266582e-05,
787
+ "loss": 0.6924,
788
+ "step": 1220
789
+ },
790
+ {
791
+ "epoch": 2.17,
792
+ "learning_rate": 8.804755944931163e-05,
793
+ "loss": 0.6956,
794
+ "step": 1230
795
+ },
796
+ {
797
+ "epoch": 2.19,
798
+ "learning_rate": 8.617021276595745e-05,
799
+ "loss": 0.6943,
800
+ "step": 1240
801
+ },
802
+ {
803
+ "epoch": 2.21,
804
+ "learning_rate": 8.429286608260324e-05,
805
+ "loss": 0.695,
806
+ "step": 1250
807
+ },
808
+ {
809
+ "epoch": 2.22,
810
+ "learning_rate": 8.241551939924905e-05,
811
+ "loss": 0.6897,
812
+ "step": 1260
813
+ },
814
+ {
815
+ "epoch": 2.24,
816
+ "learning_rate": 8.053817271589486e-05,
817
+ "loss": 0.6922,
818
+ "step": 1270
819
+ },
820
+ {
821
+ "epoch": 2.26,
822
+ "learning_rate": 7.866082603254068e-05,
823
+ "loss": 0.6912,
824
+ "step": 1280
825
+ },
826
+ {
827
+ "epoch": 2.28,
828
+ "learning_rate": 7.678347934918648e-05,
829
+ "loss": 0.6837,
830
+ "step": 1290
831
+ },
832
+ {
833
+ "epoch": 2.29,
834
+ "learning_rate": 7.490613266583228e-05,
835
+ "loss": 0.69,
836
+ "step": 1300
837
+ },
838
+ {
839
+ "epoch": 2.31,
840
+ "learning_rate": 7.30287859824781e-05,
841
+ "loss": 0.6968,
842
+ "step": 1310
843
+ },
844
+ {
845
+ "epoch": 2.33,
846
+ "learning_rate": 7.11514392991239e-05,
847
+ "loss": 0.6898,
848
+ "step": 1320
849
+ },
850
+ {
851
+ "epoch": 2.35,
852
+ "learning_rate": 6.927409261576971e-05,
853
+ "loss": 0.6906,
854
+ "step": 1330
855
+ },
856
+ {
857
+ "epoch": 2.36,
858
+ "learning_rate": 6.739674593241552e-05,
859
+ "loss": 0.6951,
860
+ "step": 1340
861
+ },
862
+ {
863
+ "epoch": 2.38,
864
+ "learning_rate": 6.551939924906132e-05,
865
+ "loss": 0.692,
866
+ "step": 1350
867
+ },
868
+ {
869
+ "epoch": 2.4,
870
+ "learning_rate": 6.364205256570713e-05,
871
+ "loss": 0.6868,
872
+ "step": 1360
873
+ },
874
+ {
875
+ "epoch": 2.42,
876
+ "learning_rate": 6.176470588235294e-05,
877
+ "loss": 0.6895,
878
+ "step": 1370
879
+ },
880
+ {
881
+ "epoch": 2.44,
882
+ "learning_rate": 5.988735919899874e-05,
883
+ "loss": 0.6937,
884
+ "step": 1380
885
+ },
886
+ {
887
+ "epoch": 2.45,
888
+ "learning_rate": 5.801001251564455e-05,
889
+ "loss": 0.6825,
890
+ "step": 1390
891
+ },
892
+ {
893
+ "epoch": 2.47,
894
+ "learning_rate": 5.6132665832290355e-05,
895
+ "loss": 0.6753,
896
+ "step": 1400
897
+ },
898
+ {
899
+ "epoch": 2.47,
900
+ "eval_loss": 0.6928293704986572,
901
+ "eval_runtime": 65.2965,
902
+ "eval_samples_per_second": 30.63,
903
+ "eval_steps_per_second": 0.766,
904
+ "step": 1400
905
+ }
906
+ ],
907
+ "max_steps": 1698,
908
+ "num_train_epochs": 3,
909
+ "total_flos": 6.472761202517737e+18,
910
+ "trial_name": null,
911
+ "trial_params": null
912
+ }
checkpoint-1400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3475910ba01356354994b8c54e7272fa9f9ab35fefffcb1f6084a7aa7bd2ffae
3
+ size 3643
checkpoint-1600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21471b74d2519eeaf82545d7409c586d4284e156469f9def83d42a8cb55e6b0e
3
+ size 52562757
checkpoint-1600/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a1b1afb254ad67e0518bc6b4518be2d93b688956669a9b4f8519b93e00445be
3
+ size 13370465717
checkpoint-1600/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ddfd0ddfe725d5bbd5c6f21d3df0ea8f1e668bd78abe0538b9f4458ccc7c454
3
+ size 14583
checkpoint-1600/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db88e995b004c739d6e5bd81b4a2b2b47807c42c040729a83c27fb6b583e85c0
3
+ size 14583
checkpoint-1600/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d01af230a3de8fe7b6073674caf2d94b6c0d5bbca9f616bece34b7ed38c2756d
3
+ size 14583
checkpoint-1600/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efaf75db5e6c6f80aed530c834f822323aa91c734eb3637693ea84d0e8d2fedb
3
+ size 14583
checkpoint-1600/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e26af6656fa5cdd60b6b7be34328643f531b957ab64ad85817a65535bc5c5ab8
3
+ size 14583
checkpoint-1600/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a27972b07b3b3794f0ed4c4540dc5628e564d64b63632beeee7439ff48d2128
3
+ size 557
checkpoint-1600/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7d3fbb2494533bb5b72a386e182f76b9b912cb864752f3170dd27d2cb72ee9d
3
+ size 627
checkpoint-1600/trainer_state.json ADDED
@@ -0,0 +1,1040 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.691353976726532,
3
+ "best_model_checkpoint": "/home/llmadmin/models/loras-trained/wizardLM-llama-lora-13b/checkpoint-1600",
4
+ "epoch": 2.8235294117647056,
5
+ "global_step": 1600,
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": 2.6999999999999996e-05,
13
+ "loss": 1.2009,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.04,
18
+ "learning_rate": 5.6999999999999996e-05,
19
+ "loss": 1.198,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.05,
24
+ "learning_rate": 8.699999999999999e-05,
25
+ "loss": 1.1318,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.07,
30
+ "learning_rate": 0.000117,
31
+ "loss": 1.0446,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.09,
36
+ "learning_rate": 0.000147,
37
+ "loss": 0.9254,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.11,
42
+ "learning_rate": 0.00017699999999999997,
43
+ "loss": 0.8364,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "learning_rate": 0.00020699999999999996,
49
+ "loss": 0.818,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.14,
54
+ "learning_rate": 0.000237,
55
+ "loss": 0.8094,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.16,
60
+ "learning_rate": 0.000267,
61
+ "loss": 0.7891,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.18,
66
+ "learning_rate": 0.00029699999999999996,
67
+ "loss": 0.7915,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.19,
72
+ "learning_rate": 0.0002983103879849812,
73
+ "loss": 0.777,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.21,
78
+ "learning_rate": 0.000296433041301627,
79
+ "loss": 0.7809,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.23,
84
+ "learning_rate": 0.0002945556946182728,
85
+ "loss": 0.7705,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.25,
90
+ "learning_rate": 0.0002926783479349186,
91
+ "loss": 0.7545,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.26,
96
+ "learning_rate": 0.00029080100125156443,
97
+ "loss": 0.7584,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.28,
102
+ "learning_rate": 0.00028892365456821025,
103
+ "loss": 0.751,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.3,
108
+ "learning_rate": 0.00028704630788485607,
109
+ "loss": 0.757,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.32,
114
+ "learning_rate": 0.00028516896120150183,
115
+ "loss": 0.7446,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.34,
120
+ "learning_rate": 0.00028329161451814765,
121
+ "loss": 0.7411,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.35,
126
+ "learning_rate": 0.0002814142678347935,
127
+ "loss": 0.754,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.35,
132
+ "eval_loss": 0.7339245676994324,
133
+ "eval_runtime": 65.2861,
134
+ "eval_samples_per_second": 30.634,
135
+ "eval_steps_per_second": 0.766,
136
+ "step": 200
137
+ },
138
+ {
139
+ "epoch": 0.37,
140
+ "learning_rate": 0.00027953692115143924,
141
+ "loss": 0.7487,
142
+ "step": 210
143
+ },
144
+ {
145
+ "epoch": 0.39,
146
+ "learning_rate": 0.00027765957446808506,
147
+ "loss": 0.7478,
148
+ "step": 220
149
+ },
150
+ {
151
+ "epoch": 0.41,
152
+ "learning_rate": 0.0002757822277847309,
153
+ "loss": 0.7336,
154
+ "step": 230
155
+ },
156
+ {
157
+ "epoch": 0.42,
158
+ "learning_rate": 0.0002739048811013767,
159
+ "loss": 0.7386,
160
+ "step": 240
161
+ },
162
+ {
163
+ "epoch": 0.44,
164
+ "learning_rate": 0.0002720275344180225,
165
+ "loss": 0.7359,
166
+ "step": 250
167
+ },
168
+ {
169
+ "epoch": 0.46,
170
+ "learning_rate": 0.0002701501877346683,
171
+ "loss": 0.7455,
172
+ "step": 260
173
+ },
174
+ {
175
+ "epoch": 0.48,
176
+ "learning_rate": 0.0002682728410513141,
177
+ "loss": 0.733,
178
+ "step": 270
179
+ },
180
+ {
181
+ "epoch": 0.49,
182
+ "learning_rate": 0.0002663954943679599,
183
+ "loss": 0.7332,
184
+ "step": 280
185
+ },
186
+ {
187
+ "epoch": 0.51,
188
+ "learning_rate": 0.00026451814768460575,
189
+ "loss": 0.7253,
190
+ "step": 290
191
+ },
192
+ {
193
+ "epoch": 0.53,
194
+ "learning_rate": 0.00026264080100125157,
195
+ "loss": 0.732,
196
+ "step": 300
197
+ },
198
+ {
199
+ "epoch": 0.55,
200
+ "learning_rate": 0.00026076345431789733,
201
+ "loss": 0.7271,
202
+ "step": 310
203
+ },
204
+ {
205
+ "epoch": 0.56,
206
+ "learning_rate": 0.00025888610763454315,
207
+ "loss": 0.7365,
208
+ "step": 320
209
+ },
210
+ {
211
+ "epoch": 0.58,
212
+ "learning_rate": 0.00025700876095118897,
213
+ "loss": 0.7266,
214
+ "step": 330
215
+ },
216
+ {
217
+ "epoch": 0.6,
218
+ "learning_rate": 0.0002551314142678348,
219
+ "loss": 0.733,
220
+ "step": 340
221
+ },
222
+ {
223
+ "epoch": 0.62,
224
+ "learning_rate": 0.00025325406758448056,
225
+ "loss": 0.7343,
226
+ "step": 350
227
+ },
228
+ {
229
+ "epoch": 0.64,
230
+ "learning_rate": 0.0002513767209011264,
231
+ "loss": 0.723,
232
+ "step": 360
233
+ },
234
+ {
235
+ "epoch": 0.65,
236
+ "learning_rate": 0.0002494993742177722,
237
+ "loss": 0.7391,
238
+ "step": 370
239
+ },
240
+ {
241
+ "epoch": 0.67,
242
+ "learning_rate": 0.000247622027534418,
243
+ "loss": 0.7265,
244
+ "step": 380
245
+ },
246
+ {
247
+ "epoch": 0.69,
248
+ "learning_rate": 0.00024574468085106384,
249
+ "loss": 0.7283,
250
+ "step": 390
251
+ },
252
+ {
253
+ "epoch": 0.71,
254
+ "learning_rate": 0.0002438673341677096,
255
+ "loss": 0.7306,
256
+ "step": 400
257
+ },
258
+ {
259
+ "epoch": 0.71,
260
+ "eval_loss": 0.7154878377914429,
261
+ "eval_runtime": 65.4382,
262
+ "eval_samples_per_second": 30.563,
263
+ "eval_steps_per_second": 0.764,
264
+ "step": 400
265
+ },
266
+ {
267
+ "epoch": 0.72,
268
+ "learning_rate": 0.00024198998748435542,
269
+ "loss": 0.725,
270
+ "step": 410
271
+ },
272
+ {
273
+ "epoch": 0.74,
274
+ "learning_rate": 0.00024011264080100122,
275
+ "loss": 0.7275,
276
+ "step": 420
277
+ },
278
+ {
279
+ "epoch": 0.76,
280
+ "learning_rate": 0.000238235294117647,
281
+ "loss": 0.7188,
282
+ "step": 430
283
+ },
284
+ {
285
+ "epoch": 0.78,
286
+ "learning_rate": 0.00023635794743429286,
287
+ "loss": 0.7216,
288
+ "step": 440
289
+ },
290
+ {
291
+ "epoch": 0.79,
292
+ "learning_rate": 0.00023448060075093865,
293
+ "loss": 0.7132,
294
+ "step": 450
295
+ },
296
+ {
297
+ "epoch": 0.81,
298
+ "learning_rate": 0.00023260325406758447,
299
+ "loss": 0.7372,
300
+ "step": 460
301
+ },
302
+ {
303
+ "epoch": 0.83,
304
+ "learning_rate": 0.00023072590738423026,
305
+ "loss": 0.7111,
306
+ "step": 470
307
+ },
308
+ {
309
+ "epoch": 0.85,
310
+ "learning_rate": 0.00022884856070087608,
311
+ "loss": 0.7131,
312
+ "step": 480
313
+ },
314
+ {
315
+ "epoch": 0.86,
316
+ "learning_rate": 0.00022697121401752188,
317
+ "loss": 0.7193,
318
+ "step": 490
319
+ },
320
+ {
321
+ "epoch": 0.88,
322
+ "learning_rate": 0.00022509386733416767,
323
+ "loss": 0.7152,
324
+ "step": 500
325
+ },
326
+ {
327
+ "epoch": 0.9,
328
+ "learning_rate": 0.00022321652065081352,
329
+ "loss": 0.7234,
330
+ "step": 510
331
+ },
332
+ {
333
+ "epoch": 0.92,
334
+ "learning_rate": 0.0002213391739674593,
335
+ "loss": 0.7193,
336
+ "step": 520
337
+ },
338
+ {
339
+ "epoch": 0.94,
340
+ "learning_rate": 0.00021946182728410513,
341
+ "loss": 0.7208,
342
+ "step": 530
343
+ },
344
+ {
345
+ "epoch": 0.95,
346
+ "learning_rate": 0.00021758448060075092,
347
+ "loss": 0.7205,
348
+ "step": 540
349
+ },
350
+ {
351
+ "epoch": 0.97,
352
+ "learning_rate": 0.00021570713391739672,
353
+ "loss": 0.7154,
354
+ "step": 550
355
+ },
356
+ {
357
+ "epoch": 0.99,
358
+ "learning_rate": 0.00021382978723404254,
359
+ "loss": 0.7224,
360
+ "step": 560
361
+ },
362
+ {
363
+ "epoch": 1.01,
364
+ "learning_rate": 0.00021195244055068833,
365
+ "loss": 0.7015,
366
+ "step": 570
367
+ },
368
+ {
369
+ "epoch": 1.02,
370
+ "learning_rate": 0.00021007509386733418,
371
+ "loss": 0.7168,
372
+ "step": 580
373
+ },
374
+ {
375
+ "epoch": 1.04,
376
+ "learning_rate": 0.00020819774718397997,
377
+ "loss": 0.7281,
378
+ "step": 590
379
+ },
380
+ {
381
+ "epoch": 1.06,
382
+ "learning_rate": 0.00020632040050062576,
383
+ "loss": 0.7112,
384
+ "step": 600
385
+ },
386
+ {
387
+ "epoch": 1.06,
388
+ "eval_loss": 0.7068994045257568,
389
+ "eval_runtime": 65.2314,
390
+ "eval_samples_per_second": 30.66,
391
+ "eval_steps_per_second": 0.767,
392
+ "step": 600
393
+ },
394
+ {
395
+ "epoch": 1.08,
396
+ "learning_rate": 0.00020444305381727158,
397
+ "loss": 0.7133,
398
+ "step": 610
399
+ },
400
+ {
401
+ "epoch": 1.09,
402
+ "learning_rate": 0.00020256570713391738,
403
+ "loss": 0.7047,
404
+ "step": 620
405
+ },
406
+ {
407
+ "epoch": 1.11,
408
+ "learning_rate": 0.00020068836045056317,
409
+ "loss": 0.7163,
410
+ "step": 630
411
+ },
412
+ {
413
+ "epoch": 1.13,
414
+ "learning_rate": 0.000198811013767209,
415
+ "loss": 0.7168,
416
+ "step": 640
417
+ },
418
+ {
419
+ "epoch": 1.15,
420
+ "learning_rate": 0.00019693366708385478,
421
+ "loss": 0.7028,
422
+ "step": 650
423
+ },
424
+ {
425
+ "epoch": 1.16,
426
+ "learning_rate": 0.00019505632040050063,
427
+ "loss": 0.7064,
428
+ "step": 660
429
+ },
430
+ {
431
+ "epoch": 1.18,
432
+ "learning_rate": 0.00019317897371714642,
433
+ "loss": 0.7127,
434
+ "step": 670
435
+ },
436
+ {
437
+ "epoch": 1.2,
438
+ "learning_rate": 0.00019130162703379222,
439
+ "loss": 0.7193,
440
+ "step": 680
441
+ },
442
+ {
443
+ "epoch": 1.22,
444
+ "learning_rate": 0.00018942428035043804,
445
+ "loss": 0.6966,
446
+ "step": 690
447
+ },
448
+ {
449
+ "epoch": 1.24,
450
+ "learning_rate": 0.00018754693366708383,
451
+ "loss": 0.7104,
452
+ "step": 700
453
+ },
454
+ {
455
+ "epoch": 1.25,
456
+ "learning_rate": 0.00018566958698372962,
457
+ "loss": 0.7076,
458
+ "step": 710
459
+ },
460
+ {
461
+ "epoch": 1.27,
462
+ "learning_rate": 0.00018379224030037544,
463
+ "loss": 0.6971,
464
+ "step": 720
465
+ },
466
+ {
467
+ "epoch": 1.29,
468
+ "learning_rate": 0.00018191489361702126,
469
+ "loss": 0.6941,
470
+ "step": 730
471
+ },
472
+ {
473
+ "epoch": 1.31,
474
+ "learning_rate": 0.00018003754693366708,
475
+ "loss": 0.6992,
476
+ "step": 740
477
+ },
478
+ {
479
+ "epoch": 1.32,
480
+ "learning_rate": 0.00017816020025031287,
481
+ "loss": 0.7059,
482
+ "step": 750
483
+ },
484
+ {
485
+ "epoch": 1.34,
486
+ "learning_rate": 0.0001762828535669587,
487
+ "loss": 0.6936,
488
+ "step": 760
489
+ },
490
+ {
491
+ "epoch": 1.36,
492
+ "learning_rate": 0.0001744055068836045,
493
+ "loss": 0.6961,
494
+ "step": 770
495
+ },
496
+ {
497
+ "epoch": 1.38,
498
+ "learning_rate": 0.00017252816020025028,
499
+ "loss": 0.7125,
500
+ "step": 780
501
+ },
502
+ {
503
+ "epoch": 1.39,
504
+ "learning_rate": 0.0001706508135168961,
505
+ "loss": 0.7046,
506
+ "step": 790
507
+ },
508
+ {
509
+ "epoch": 1.41,
510
+ "learning_rate": 0.00016877346683354192,
511
+ "loss": 0.7034,
512
+ "step": 800
513
+ },
514
+ {
515
+ "epoch": 1.41,
516
+ "eval_loss": 0.7016428709030151,
517
+ "eval_runtime": 65.2245,
518
+ "eval_samples_per_second": 30.663,
519
+ "eval_steps_per_second": 0.767,
520
+ "step": 800
521
+ },
522
+ {
523
+ "epoch": 1.43,
524
+ "learning_rate": 0.00016689612015018774,
525
+ "loss": 0.7011,
526
+ "step": 810
527
+ },
528
+ {
529
+ "epoch": 1.45,
530
+ "learning_rate": 0.00016501877346683353,
531
+ "loss": 0.7049,
532
+ "step": 820
533
+ },
534
+ {
535
+ "epoch": 1.46,
536
+ "learning_rate": 0.00016314142678347933,
537
+ "loss": 0.6965,
538
+ "step": 830
539
+ },
540
+ {
541
+ "epoch": 1.48,
542
+ "learning_rate": 0.00016126408010012515,
543
+ "loss": 0.6965,
544
+ "step": 840
545
+ },
546
+ {
547
+ "epoch": 1.5,
548
+ "learning_rate": 0.00015938673341677094,
549
+ "loss": 0.6975,
550
+ "step": 850
551
+ },
552
+ {
553
+ "epoch": 1.52,
554
+ "learning_rate": 0.00015750938673341673,
555
+ "loss": 0.7094,
556
+ "step": 860
557
+ },
558
+ {
559
+ "epoch": 1.54,
560
+ "learning_rate": 0.00015563204005006258,
561
+ "loss": 0.6918,
562
+ "step": 870
563
+ },
564
+ {
565
+ "epoch": 1.55,
566
+ "learning_rate": 0.00015375469336670837,
567
+ "loss": 0.7031,
568
+ "step": 880
569
+ },
570
+ {
571
+ "epoch": 1.57,
572
+ "learning_rate": 0.0001518773466833542,
573
+ "loss": 0.7027,
574
+ "step": 890
575
+ },
576
+ {
577
+ "epoch": 1.59,
578
+ "learning_rate": 0.00015,
579
+ "loss": 0.6996,
580
+ "step": 900
581
+ },
582
+ {
583
+ "epoch": 1.61,
584
+ "learning_rate": 0.00014812265331664578,
585
+ "loss": 0.7048,
586
+ "step": 910
587
+ },
588
+ {
589
+ "epoch": 1.62,
590
+ "learning_rate": 0.0001462453066332916,
591
+ "loss": 0.6962,
592
+ "step": 920
593
+ },
594
+ {
595
+ "epoch": 1.64,
596
+ "learning_rate": 0.00014436795994993742,
597
+ "loss": 0.6918,
598
+ "step": 930
599
+ },
600
+ {
601
+ "epoch": 1.66,
602
+ "learning_rate": 0.0001424906132665832,
603
+ "loss": 0.6979,
604
+ "step": 940
605
+ },
606
+ {
607
+ "epoch": 1.68,
608
+ "learning_rate": 0.000140613266583229,
609
+ "loss": 0.6958,
610
+ "step": 950
611
+ },
612
+ {
613
+ "epoch": 1.69,
614
+ "learning_rate": 0.00013873591989987483,
615
+ "loss": 0.7027,
616
+ "step": 960
617
+ },
618
+ {
619
+ "epoch": 1.71,
620
+ "learning_rate": 0.00013685857321652065,
621
+ "loss": 0.7092,
622
+ "step": 970
623
+ },
624
+ {
625
+ "epoch": 1.73,
626
+ "learning_rate": 0.00013498122653316644,
627
+ "loss": 0.6917,
628
+ "step": 980
629
+ },
630
+ {
631
+ "epoch": 1.75,
632
+ "learning_rate": 0.00013310387984981226,
633
+ "loss": 0.6996,
634
+ "step": 990
635
+ },
636
+ {
637
+ "epoch": 1.76,
638
+ "learning_rate": 0.00013122653316645805,
639
+ "loss": 0.6889,
640
+ "step": 1000
641
+ },
642
+ {
643
+ "epoch": 1.76,
644
+ "eval_loss": 0.6975076794624329,
645
+ "eval_runtime": 65.2345,
646
+ "eval_samples_per_second": 30.659,
647
+ "eval_steps_per_second": 0.766,
648
+ "step": 1000
649
+ },
650
+ {
651
+ "epoch": 1.78,
652
+ "learning_rate": 0.00012934918648310387,
653
+ "loss": 0.7083,
654
+ "step": 1010
655
+ },
656
+ {
657
+ "epoch": 1.8,
658
+ "learning_rate": 0.00012747183979974967,
659
+ "loss": 0.6934,
660
+ "step": 1020
661
+ },
662
+ {
663
+ "epoch": 1.82,
664
+ "learning_rate": 0.00012559449311639549,
665
+ "loss": 0.7002,
666
+ "step": 1030
667
+ },
668
+ {
669
+ "epoch": 1.84,
670
+ "learning_rate": 0.00012371714643304128,
671
+ "loss": 0.6899,
672
+ "step": 1040
673
+ },
674
+ {
675
+ "epoch": 1.85,
676
+ "learning_rate": 0.0001218397997496871,
677
+ "loss": 0.6856,
678
+ "step": 1050
679
+ },
680
+ {
681
+ "epoch": 1.87,
682
+ "learning_rate": 0.00011996245306633289,
683
+ "loss": 0.695,
684
+ "step": 1060
685
+ },
686
+ {
687
+ "epoch": 1.89,
688
+ "learning_rate": 0.00011808510638297871,
689
+ "loss": 0.6934,
690
+ "step": 1070
691
+ },
692
+ {
693
+ "epoch": 1.91,
694
+ "learning_rate": 0.00011620775969962452,
695
+ "loss": 0.6933,
696
+ "step": 1080
697
+ },
698
+ {
699
+ "epoch": 1.92,
700
+ "learning_rate": 0.00011433041301627033,
701
+ "loss": 0.6983,
702
+ "step": 1090
703
+ },
704
+ {
705
+ "epoch": 1.94,
706
+ "learning_rate": 0.00011245306633291615,
707
+ "loss": 0.705,
708
+ "step": 1100
709
+ },
710
+ {
711
+ "epoch": 1.96,
712
+ "learning_rate": 0.00011057571964956194,
713
+ "loss": 0.6944,
714
+ "step": 1110
715
+ },
716
+ {
717
+ "epoch": 1.98,
718
+ "learning_rate": 0.00010869837296620774,
719
+ "loss": 0.6977,
720
+ "step": 1120
721
+ },
722
+ {
723
+ "epoch": 1.99,
724
+ "learning_rate": 0.00010682102628285355,
725
+ "loss": 0.6966,
726
+ "step": 1130
727
+ },
728
+ {
729
+ "epoch": 2.01,
730
+ "learning_rate": 0.00010494367959949937,
731
+ "loss": 0.6868,
732
+ "step": 1140
733
+ },
734
+ {
735
+ "epoch": 2.03,
736
+ "learning_rate": 0.00010306633291614518,
737
+ "loss": 0.6802,
738
+ "step": 1150
739
+ },
740
+ {
741
+ "epoch": 2.05,
742
+ "learning_rate": 0.00010118898623279097,
743
+ "loss": 0.6916,
744
+ "step": 1160
745
+ },
746
+ {
747
+ "epoch": 2.06,
748
+ "learning_rate": 9.931163954943679e-05,
749
+ "loss": 0.7018,
750
+ "step": 1170
751
+ },
752
+ {
753
+ "epoch": 2.08,
754
+ "learning_rate": 9.74342928660826e-05,
755
+ "loss": 0.6951,
756
+ "step": 1180
757
+ },
758
+ {
759
+ "epoch": 2.1,
760
+ "learning_rate": 9.55569461827284e-05,
761
+ "loss": 0.6906,
762
+ "step": 1190
763
+ },
764
+ {
765
+ "epoch": 2.12,
766
+ "learning_rate": 9.36795994993742e-05,
767
+ "loss": 0.6965,
768
+ "step": 1200
769
+ },
770
+ {
771
+ "epoch": 2.12,
772
+ "eval_loss": 0.6952534317970276,
773
+ "eval_runtime": 65.233,
774
+ "eval_samples_per_second": 30.659,
775
+ "eval_steps_per_second": 0.766,
776
+ "step": 1200
777
+ },
778
+ {
779
+ "epoch": 2.14,
780
+ "learning_rate": 9.180225281602002e-05,
781
+ "loss": 0.6969,
782
+ "step": 1210
783
+ },
784
+ {
785
+ "epoch": 2.15,
786
+ "learning_rate": 8.992490613266582e-05,
787
+ "loss": 0.6924,
788
+ "step": 1220
789
+ },
790
+ {
791
+ "epoch": 2.17,
792
+ "learning_rate": 8.804755944931163e-05,
793
+ "loss": 0.6956,
794
+ "step": 1230
795
+ },
796
+ {
797
+ "epoch": 2.19,
798
+ "learning_rate": 8.617021276595745e-05,
799
+ "loss": 0.6943,
800
+ "step": 1240
801
+ },
802
+ {
803
+ "epoch": 2.21,
804
+ "learning_rate": 8.429286608260324e-05,
805
+ "loss": 0.695,
806
+ "step": 1250
807
+ },
808
+ {
809
+ "epoch": 2.22,
810
+ "learning_rate": 8.241551939924905e-05,
811
+ "loss": 0.6897,
812
+ "step": 1260
813
+ },
814
+ {
815
+ "epoch": 2.24,
816
+ "learning_rate": 8.053817271589486e-05,
817
+ "loss": 0.6922,
818
+ "step": 1270
819
+ },
820
+ {
821
+ "epoch": 2.26,
822
+ "learning_rate": 7.866082603254068e-05,
823
+ "loss": 0.6912,
824
+ "step": 1280
825
+ },
826
+ {
827
+ "epoch": 2.28,
828
+ "learning_rate": 7.678347934918648e-05,
829
+ "loss": 0.6837,
830
+ "step": 1290
831
+ },
832
+ {
833
+ "epoch": 2.29,
834
+ "learning_rate": 7.490613266583228e-05,
835
+ "loss": 0.69,
836
+ "step": 1300
837
+ },
838
+ {
839
+ "epoch": 2.31,
840
+ "learning_rate": 7.30287859824781e-05,
841
+ "loss": 0.6968,
842
+ "step": 1310
843
+ },
844
+ {
845
+ "epoch": 2.33,
846
+ "learning_rate": 7.11514392991239e-05,
847
+ "loss": 0.6898,
848
+ "step": 1320
849
+ },
850
+ {
851
+ "epoch": 2.35,
852
+ "learning_rate": 6.927409261576971e-05,
853
+ "loss": 0.6906,
854
+ "step": 1330
855
+ },
856
+ {
857
+ "epoch": 2.36,
858
+ "learning_rate": 6.739674593241552e-05,
859
+ "loss": 0.6951,
860
+ "step": 1340
861
+ },
862
+ {
863
+ "epoch": 2.38,
864
+ "learning_rate": 6.551939924906132e-05,
865
+ "loss": 0.692,
866
+ "step": 1350
867
+ },
868
+ {
869
+ "epoch": 2.4,
870
+ "learning_rate": 6.364205256570713e-05,
871
+ "loss": 0.6868,
872
+ "step": 1360
873
+ },
874
+ {
875
+ "epoch": 2.42,
876
+ "learning_rate": 6.176470588235294e-05,
877
+ "loss": 0.6895,
878
+ "step": 1370
879
+ },
880
+ {
881
+ "epoch": 2.44,
882
+ "learning_rate": 5.988735919899874e-05,
883
+ "loss": 0.6937,
884
+ "step": 1380
885
+ },
886
+ {
887
+ "epoch": 2.45,
888
+ "learning_rate": 5.801001251564455e-05,
889
+ "loss": 0.6825,
890
+ "step": 1390
891
+ },
892
+ {
893
+ "epoch": 2.47,
894
+ "learning_rate": 5.6132665832290355e-05,
895
+ "loss": 0.6753,
896
+ "step": 1400
897
+ },
898
+ {
899
+ "epoch": 2.47,
900
+ "eval_loss": 0.6928293704986572,
901
+ "eval_runtime": 65.2965,
902
+ "eval_samples_per_second": 30.63,
903
+ "eval_steps_per_second": 0.766,
904
+ "step": 1400
905
+ },
906
+ {
907
+ "epoch": 2.49,
908
+ "learning_rate": 5.425531914893617e-05,
909
+ "loss": 0.7009,
910
+ "step": 1410
911
+ },
912
+ {
913
+ "epoch": 2.51,
914
+ "learning_rate": 5.237797246558197e-05,
915
+ "loss": 0.6943,
916
+ "step": 1420
917
+ },
918
+ {
919
+ "epoch": 2.52,
920
+ "learning_rate": 5.050062578222778e-05,
921
+ "loss": 0.6957,
922
+ "step": 1430
923
+ },
924
+ {
925
+ "epoch": 2.54,
926
+ "learning_rate": 4.862327909887359e-05,
927
+ "loss": 0.6813,
928
+ "step": 1440
929
+ },
930
+ {
931
+ "epoch": 2.56,
932
+ "learning_rate": 4.6745932415519395e-05,
933
+ "loss": 0.683,
934
+ "step": 1450
935
+ },
936
+ {
937
+ "epoch": 2.58,
938
+ "learning_rate": 4.48685857321652e-05,
939
+ "loss": 0.6946,
940
+ "step": 1460
941
+ },
942
+ {
943
+ "epoch": 2.59,
944
+ "learning_rate": 4.299123904881101e-05,
945
+ "loss": 0.6843,
946
+ "step": 1470
947
+ },
948
+ {
949
+ "epoch": 2.61,
950
+ "learning_rate": 4.111389236545682e-05,
951
+ "loss": 0.6919,
952
+ "step": 1480
953
+ },
954
+ {
955
+ "epoch": 2.63,
956
+ "learning_rate": 3.923654568210262e-05,
957
+ "loss": 0.6887,
958
+ "step": 1490
959
+ },
960
+ {
961
+ "epoch": 2.65,
962
+ "learning_rate": 3.735919899874843e-05,
963
+ "loss": 0.6865,
964
+ "step": 1500
965
+ },
966
+ {
967
+ "epoch": 2.66,
968
+ "learning_rate": 3.548185231539424e-05,
969
+ "loss": 0.6858,
970
+ "step": 1510
971
+ },
972
+ {
973
+ "epoch": 2.68,
974
+ "learning_rate": 3.360450563204005e-05,
975
+ "loss": 0.6907,
976
+ "step": 1520
977
+ },
978
+ {
979
+ "epoch": 2.7,
980
+ "learning_rate": 3.1727158948685854e-05,
981
+ "loss": 0.6841,
982
+ "step": 1530
983
+ },
984
+ {
985
+ "epoch": 2.72,
986
+ "learning_rate": 2.9849812265331664e-05,
987
+ "loss": 0.6901,
988
+ "step": 1540
989
+ },
990
+ {
991
+ "epoch": 2.74,
992
+ "learning_rate": 2.797246558197747e-05,
993
+ "loss": 0.6912,
994
+ "step": 1550
995
+ },
996
+ {
997
+ "epoch": 2.75,
998
+ "learning_rate": 2.6095118898623277e-05,
999
+ "loss": 0.6893,
1000
+ "step": 1560
1001
+ },
1002
+ {
1003
+ "epoch": 2.77,
1004
+ "learning_rate": 2.4217772215269084e-05,
1005
+ "loss": 0.6959,
1006
+ "step": 1570
1007
+ },
1008
+ {
1009
+ "epoch": 2.79,
1010
+ "learning_rate": 2.234042553191489e-05,
1011
+ "loss": 0.6835,
1012
+ "step": 1580
1013
+ },
1014
+ {
1015
+ "epoch": 2.81,
1016
+ "learning_rate": 2.0463078848560697e-05,
1017
+ "loss": 0.6855,
1018
+ "step": 1590
1019
+ },
1020
+ {
1021
+ "epoch": 2.82,
1022
+ "learning_rate": 1.8585732165206507e-05,
1023
+ "loss": 0.6935,
1024
+ "step": 1600
1025
+ },
1026
+ {
1027
+ "epoch": 2.82,
1028
+ "eval_loss": 0.691353976726532,
1029
+ "eval_runtime": 65.2035,
1030
+ "eval_samples_per_second": 30.673,
1031
+ "eval_steps_per_second": 0.767,
1032
+ "step": 1600
1033
+ }
1034
+ ],
1035
+ "max_steps": 1698,
1036
+ "num_train_epochs": 3,
1037
+ "total_flos": 7.398050488408605e+18,
1038
+ "trial_name": null,
1039
+ "trial_params": null
1040
+ }
checkpoint-1600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3475910ba01356354994b8c54e7272fa9f9ab35fefffcb1f6084a7aa7bd2ffae
3
+ size 3643