ytcheng commited on
Commit
0102256
·
verified ·
1 Parent(s): af8446a

End of training

Browse files
README.md CHANGED
@@ -2,6 +2,7 @@
2
  library_name: peft
3
  tags:
4
  - llama-factory
 
5
  - generated_from_trainer
6
  base_model: Qwen/Qwen-14B
7
  model-index:
@@ -14,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  # Qwen-14B-lora-pretrain
16
 
17
- This model is a fine-tuned version of [Qwen/Qwen-14B](https://huggingface.co/Qwen/Qwen-14B) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 2.0424
20
 
21
  ## Model description
22
 
 
2
  library_name: peft
3
  tags:
4
  - llama-factory
5
+ - lora
6
  - generated_from_trainer
7
  base_model: Qwen/Qwen-14B
8
  model-index:
 
15
 
16
  # Qwen-14B-lora-pretrain
17
 
18
+ This model is a fine-tuned version of [Qwen/Qwen-14B](https://huggingface.co/Qwen/Qwen-14B) on the sm_artile dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 2.0396
21
 
22
  ## Model description
23
 
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.998264893001735,
3
+ "eval_loss": 2.0396180152893066,
4
+ "eval_runtime": 154.6591,
5
+ "eval_samples_per_second": 3.731,
6
+ "eval_steps_per_second": 3.731,
7
+ "perplexity": 7.687672064754986,
8
+ "total_flos": 1.2799386304118784e+18,
9
+ "train_loss": 2.14731617155389,
10
+ "train_runtime": 14043.9874,
11
+ "train_samples_per_second": 1.108,
12
+ "train_steps_per_second": 0.138
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.998264893001735,
3
+ "eval_loss": 2.0396180152893066,
4
+ "eval_runtime": 154.6591,
5
+ "eval_samples_per_second": 3.731,
6
+ "eval_steps_per_second": 3.731,
7
+ "perplexity": 7.687672064754986
8
+ }
runs/May11_02-40-27_d96183b7b29f/events.out.tfevents.1715409508.d96183b7b29f.1272.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f442a93598369873a111963abbc8182517cd67c14aa63f180807b060d3a15e2d
3
+ size 359
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.998264893001735,
3
+ "total_flos": 1.2799386304118784e+18,
4
+ "train_loss": 2.14731617155389,
5
+ "train_runtime": 14043.9874,
6
+ "train_samples_per_second": 1.108,
7
+ "train_steps_per_second": 0.138
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1412 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.998264893001735,
5
+ "eval_steps": 500,
6
+ "global_step": 1944,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01542317331791016,
13
+ "grad_norm": 0.06286442279815674,
14
+ "learning_rate": 5e-06,
15
+ "loss": 2.7232,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.03084634663582032,
20
+ "grad_norm": 0.06427137553691864,
21
+ "learning_rate": 1e-05,
22
+ "loss": 2.8025,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.04626951995373048,
27
+ "grad_norm": 0.17838376760482788,
28
+ "learning_rate": 1.5e-05,
29
+ "loss": 2.7466,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.06169269327164064,
34
+ "grad_norm": 0.07409191876649857,
35
+ "learning_rate": 2e-05,
36
+ "loss": 2.7862,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.0771158665895508,
41
+ "grad_norm": 0.08752360939979553,
42
+ "learning_rate": 2.5e-05,
43
+ "loss": 2.7836,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.09253903990746096,
48
+ "grad_norm": 0.11069338023662567,
49
+ "learning_rate": 3e-05,
50
+ "loss": 2.7271,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.10796221322537113,
55
+ "grad_norm": 0.1447472721338272,
56
+ "learning_rate": 3.5e-05,
57
+ "loss": 2.7783,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.12338538654328128,
62
+ "grad_norm": 0.12579452991485596,
63
+ "learning_rate": 4e-05,
64
+ "loss": 2.7324,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.13880855986119145,
69
+ "grad_norm": 0.16387595236301422,
70
+ "learning_rate": 4.5e-05,
71
+ "loss": 2.7053,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.1542317331791016,
76
+ "grad_norm": 0.19016346335411072,
77
+ "learning_rate": 5e-05,
78
+ "loss": 2.5806,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.16965490649701176,
83
+ "grad_norm": 0.21213597059249878,
84
+ "learning_rate": 5.500000000000001e-05,
85
+ "loss": 2.7242,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.18507807981492191,
90
+ "grad_norm": 0.1880221664905548,
91
+ "learning_rate": 6e-05,
92
+ "loss": 2.6331,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.20050125313283207,
97
+ "grad_norm": 0.7994162440299988,
98
+ "learning_rate": 6.500000000000001e-05,
99
+ "loss": 2.508,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.21592442645074225,
104
+ "grad_norm": 0.48757001757621765,
105
+ "learning_rate": 7e-05,
106
+ "loss": 2.4189,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.2313475997686524,
111
+ "grad_norm": 0.6417515277862549,
112
+ "learning_rate": 7.500000000000001e-05,
113
+ "loss": 2.351,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.24677077308656256,
118
+ "grad_norm": 0.45560503005981445,
119
+ "learning_rate": 8e-05,
120
+ "loss": 2.2376,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.26219394640447274,
125
+ "grad_norm": 0.43604227900505066,
126
+ "learning_rate": 8.5e-05,
127
+ "loss": 2.2678,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.2776171197223829,
132
+ "grad_norm": 0.3299252986907959,
133
+ "learning_rate": 9e-05,
134
+ "loss": 2.2033,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.29304029304029305,
139
+ "grad_norm": 0.3385520279407501,
140
+ "learning_rate": 9.5e-05,
141
+ "loss": 2.126,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.3084634663582032,
146
+ "grad_norm": 0.380649209022522,
147
+ "learning_rate": 0.0001,
148
+ "loss": 2.4127,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.32388663967611336,
153
+ "grad_norm": 0.4573318064212799,
154
+ "learning_rate": 9.999188786725007e-05,
155
+ "loss": 2.0718,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.3393098129940235,
160
+ "grad_norm": 0.5013952255249023,
161
+ "learning_rate": 9.996755410126815e-05,
162
+ "loss": 2.2014,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.3547329863119337,
167
+ "grad_norm": 0.3462372422218323,
168
+ "learning_rate": 9.992700659800388e-05,
169
+ "loss": 2.0259,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.37015615962984383,
174
+ "grad_norm": 0.3076546788215637,
175
+ "learning_rate": 9.987025851452639e-05,
176
+ "loss": 2.3585,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.385579332947754,
181
+ "grad_norm": 0.36271095275878906,
182
+ "learning_rate": 9.979732826475515e-05,
183
+ "loss": 2.2215,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.40100250626566414,
188
+ "grad_norm": 0.3243629038333893,
189
+ "learning_rate": 9.970823951348487e-05,
190
+ "loss": 2.3239,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.4164256795835743,
195
+ "grad_norm": 0.3210390508174896,
196
+ "learning_rate": 9.960302116870661e-05,
197
+ "loss": 2.2053,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.4318488529014845,
202
+ "grad_norm": 0.5260313749313354,
203
+ "learning_rate": 9.948170737222762e-05,
204
+ "loss": 2.2688,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.44727202621939466,
209
+ "grad_norm": 0.3700290024280548,
210
+ "learning_rate": 9.934433748859274e-05,
211
+ "loss": 2.2533,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.4626951995373048,
216
+ "grad_norm": 0.4328364431858063,
217
+ "learning_rate": 9.919095609231126e-05,
218
+ "loss": 2.1101,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.47811837285521497,
223
+ "grad_norm": 0.3829161524772644,
224
+ "learning_rate": 9.902161295339307e-05,
225
+ "loss": 2.1293,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.4935415461731251,
230
+ "grad_norm": 0.43613630533218384,
231
+ "learning_rate": 9.883636302119912e-05,
232
+ "loss": 2.0417,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.5089647194910353,
237
+ "grad_norm": 0.3980802595615387,
238
+ "learning_rate": 9.863526640661107e-05,
239
+ "loss": 2.1096,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.5243878928089455,
244
+ "grad_norm": 0.44878003001213074,
245
+ "learning_rate": 9.841838836252627e-05,
246
+ "loss": 2.2787,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.5398110661268556,
251
+ "grad_norm": 0.3995887339115143,
252
+ "learning_rate": 9.818579926268405e-05,
253
+ "loss": 2.3788,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.5552342394447658,
258
+ "grad_norm": 0.3169827461242676,
259
+ "learning_rate": 9.793757457883062e-05,
260
+ "loss": 2.1264,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.5706574127626759,
265
+ "grad_norm": 0.28045836091041565,
266
+ "learning_rate": 9.767379485622943e-05,
267
+ "loss": 2.2351,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.5860805860805861,
272
+ "grad_norm": 0.3078824281692505,
273
+ "learning_rate": 9.739454568752556e-05,
274
+ "loss": 2.1219,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.6015037593984962,
279
+ "grad_norm": 2.457259178161621,
280
+ "learning_rate": 9.709991768497208e-05,
281
+ "loss": 2.0565,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.6169269327164064,
286
+ "grad_norm": 0.3353936970233917,
287
+ "learning_rate": 9.679000645102771e-05,
288
+ "loss": 2.2406,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.6323501060343165,
293
+ "grad_norm": 0.4184532165527344,
294
+ "learning_rate": 9.646491254733532e-05,
295
+ "loss": 2.154,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.6477732793522267,
300
+ "grad_norm": 0.4148164987564087,
301
+ "learning_rate": 9.612474146209096e-05,
302
+ "loss": 2.1421,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.6631964526701368,
307
+ "grad_norm": 0.3182518780231476,
308
+ "learning_rate": 9.576960357581475e-05,
309
+ "loss": 2.1537,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.678619625988047,
314
+ "grad_norm": 0.3275531828403473,
315
+ "learning_rate": 9.539961412553375e-05,
316
+ "loss": 2.1743,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.6940427993059572,
321
+ "grad_norm": 0.31316521763801575,
322
+ "learning_rate": 9.501489316738945e-05,
323
+ "loss": 2.0946,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.7094659726238673,
328
+ "grad_norm": 0.3164524734020233,
329
+ "learning_rate": 9.461556553768123e-05,
330
+ "loss": 2.1048,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.7248891459417776,
335
+ "grad_norm": 0.29031479358673096,
336
+ "learning_rate": 9.420176081235881e-05,
337
+ "loss": 2.1311,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.7403123192596877,
342
+ "grad_norm": 0.3423845171928406,
343
+ "learning_rate": 9.377361326497674e-05,
344
+ "loss": 2.3041,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.7557354925775979,
349
+ "grad_norm": 0.4182330071926117,
350
+ "learning_rate": 9.333126182312465e-05,
351
+ "loss": 2.0653,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.771158665895508,
356
+ "grad_norm": 0.3934147357940674,
357
+ "learning_rate": 9.287485002334733e-05,
358
+ "loss": 1.9974,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.771158665895508,
363
+ "eval_loss": 2.0961015224456787,
364
+ "eval_runtime": 154.5123,
365
+ "eval_samples_per_second": 3.734,
366
+ "eval_steps_per_second": 3.734,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 0.7865818392134182,
371
+ "grad_norm": 0.37575840950012207,
372
+ "learning_rate": 9.240452596456905e-05,
373
+ "loss": 1.7869,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 0.8020050125313283,
378
+ "grad_norm": 0.5916942358016968,
379
+ "learning_rate": 9.192044226003789e-05,
380
+ "loss": 2.1779,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 0.8174281858492385,
385
+ "grad_norm": 0.3484373092651367,
386
+ "learning_rate": 9.142275598780473e-05,
387
+ "loss": 2.32,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 0.8328513591671486,
392
+ "grad_norm": 0.39120087027549744,
393
+ "learning_rate": 9.091162863975388e-05,
394
+ "loss": 1.9688,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 0.8482745324850588,
399
+ "grad_norm": 0.6257333755493164,
400
+ "learning_rate": 9.03872260692014e-05,
401
+ "loss": 2.231,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 0.863697705802969,
406
+ "grad_norm": 0.4078841209411621,
407
+ "learning_rate": 8.984971843707788e-05,
408
+ "loss": 2.1618,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 0.8791208791208791,
413
+ "grad_norm": 0.3387174606323242,
414
+ "learning_rate": 8.9299280156714e-05,
415
+ "loss": 2.2796,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 0.8945440524387893,
420
+ "grad_norm": 0.3817114233970642,
421
+ "learning_rate": 8.87360898372458e-05,
422
+ "loss": 1.9383,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 0.9099672257566994,
427
+ "grad_norm": 0.3814316391944885,
428
+ "learning_rate": 8.816033022565864e-05,
429
+ "loss": 2.1884,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 0.9253903990746096,
434
+ "grad_norm": 0.3003975450992584,
435
+ "learning_rate": 8.75721881474886e-05,
436
+ "loss": 2.1796,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 0.9408135723925197,
441
+ "grad_norm": 0.4703477919101715,
442
+ "learning_rate": 8.697185444620024e-05,
443
+ "loss": 2.1362,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 0.9562367457104299,
448
+ "grad_norm": 0.37337568402290344,
449
+ "learning_rate": 8.635952392126072e-05,
450
+ "loss": 2.239,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 0.97165991902834,
455
+ "grad_norm": 0.3282889127731323,
456
+ "learning_rate": 8.573539526493023e-05,
457
+ "loss": 2.15,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 0.9870830923462502,
462
+ "grad_norm": 0.3719846308231354,
463
+ "learning_rate": 8.509967099778933e-05,
464
+ "loss": 2.0602,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.0025062656641603,
469
+ "grad_norm": 0.360990434885025,
470
+ "learning_rate": 8.445255740302391e-05,
471
+ "loss": 2.228,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.0179294389820706,
476
+ "grad_norm": 0.3433365225791931,
477
+ "learning_rate": 8.379426445948932e-05,
478
+ "loss": 2.1467,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.0333526122999808,
483
+ "grad_norm": 0.33261704444885254,
484
+ "learning_rate": 8.312500577357547e-05,
485
+ "loss": 2.2814,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.048775785617891,
490
+ "grad_norm": 0.41563576459884644,
491
+ "learning_rate": 8.244499850989452e-05,
492
+ "loss": 2.046,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.064198958935801,
497
+ "grad_norm": 0.44597044587135315,
498
+ "learning_rate": 8.17544633208142e-05,
499
+ "loss": 1.9755,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.0796221322537112,
504
+ "grad_norm": 0.37877991795539856,
505
+ "learning_rate": 8.105362427485942e-05,
506
+ "loss": 2.0912,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.0950453055716214,
511
+ "grad_norm": 0.392952561378479,
512
+ "learning_rate": 8.034270878400528e-05,
513
+ "loss": 2.2183,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.1104684788895316,
518
+ "grad_norm": 0.3665911853313446,
519
+ "learning_rate": 7.962194752988519e-05,
520
+ "loss": 2.1854,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.1258916522074416,
525
+ "grad_norm": 0.3779922127723694,
526
+ "learning_rate": 7.889157438893812e-05,
527
+ "loss": 2.1174,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.1413148255253518,
532
+ "grad_norm": 0.4270557463169098,
533
+ "learning_rate": 7.815182635651913e-05,
534
+ "loss": 2.0819,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.156737998843262,
539
+ "grad_norm": 0.4602227807044983,
540
+ "learning_rate": 7.740294346999786e-05,
541
+ "loss": 2.1205,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.1721611721611722,
546
+ "grad_norm": 0.4194934666156769,
547
+ "learning_rate": 7.664516873086987e-05,
548
+ "loss": 2.1829,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.1875843454790824,
553
+ "grad_norm": 0.3483639359474182,
554
+ "learning_rate": 7.58787480259063e-05,
555
+ "loss": 2.3507,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.2030075187969924,
560
+ "grad_norm": 0.4419120252132416,
561
+ "learning_rate": 7.510393004736722e-05,
562
+ "loss": 2.1003,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.2184306921149026,
567
+ "grad_norm": 0.4768407344818115,
568
+ "learning_rate": 7.432096621230455e-05,
569
+ "loss": 2.0724,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.2338538654328128,
574
+ "grad_norm": 0.40922412276268005,
575
+ "learning_rate": 7.353011058098104e-05,
576
+ "loss": 2.0352,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.249277038750723,
581
+ "grad_norm": 0.38617295026779175,
582
+ "learning_rate": 7.273161977443137e-05,
583
+ "loss": 2.1549,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.2647002120686333,
588
+ "grad_norm": 0.3819389343261719,
589
+ "learning_rate": 7.192575289119245e-05,
590
+ "loss": 2.0827,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.2801233853865432,
595
+ "grad_norm": 0.4457474946975708,
596
+ "learning_rate": 7.11127714232297e-05,
597
+ "loss": 1.9096,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.2955465587044535,
602
+ "grad_norm": 0.47230038046836853,
603
+ "learning_rate": 7.029293917108678e-05,
604
+ "loss": 2.0398,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.3109697320223637,
609
+ "grad_norm": 0.43403419852256775,
610
+ "learning_rate": 6.946652215828618e-05,
611
+ "loss": 2.1837,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.3263929053402737,
616
+ "grad_norm": 0.3847104012966156,
617
+ "learning_rate": 6.863378854500846e-05,
618
+ "loss": 2.1997,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.3418160786581839,
623
+ "grad_norm": 0.40409407019615173,
624
+ "learning_rate": 6.779500854107827e-05,
625
+ "loss": 1.8728,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.357239251976094,
630
+ "grad_norm": 0.3734535574913025,
631
+ "learning_rate": 6.695045431828524e-05,
632
+ "loss": 2.0045,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.3726624252940043,
637
+ "grad_norm": 0.40710312128067017,
638
+ "learning_rate": 6.610039992206814e-05,
639
+ "loss": 2.1053,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 1.3880855986119145,
644
+ "grad_norm": 0.3717002868652344,
645
+ "learning_rate": 6.524512118259121e-05,
646
+ "loss": 2.1408,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 1.4035087719298245,
651
+ "grad_norm": 0.3799400329589844,
652
+ "learning_rate": 6.438489562524142e-05,
653
+ "loss": 2.242,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 1.4189319452477347,
658
+ "grad_norm": 0.3500377833843231,
659
+ "learning_rate": 6.35200023805754e-05,
660
+ "loss": 2.1352,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 1.434355118565645,
665
+ "grad_norm": 0.4466979503631592,
666
+ "learning_rate": 6.265072209374574e-05,
667
+ "loss": 2.1492,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 1.4497782918835551,
672
+ "grad_norm": 0.42561647295951843,
673
+ "learning_rate": 6.177733683343578e-05,
674
+ "loss": 2.1381,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 1.4652014652014653,
679
+ "grad_norm": 0.3749467134475708,
680
+ "learning_rate": 6.090013000033251e-05,
681
+ "loss": 2.2012,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 1.4806246385193753,
686
+ "grad_norm": 0.4494505822658539,
687
+ "learning_rate": 6.0019386235167055e-05,
688
+ "loss": 1.82,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 1.4960478118372855,
693
+ "grad_norm": 0.3980197608470917,
694
+ "learning_rate": 5.9135391326353085e-05,
695
+ "loss": 2.2977,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 1.5114709851551957,
700
+ "grad_norm": 0.4605059027671814,
701
+ "learning_rate": 5.824843211725265e-05,
702
+ "loss": 2.1393,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 1.5268941584731057,
707
+ "grad_norm": 0.4184533655643463,
708
+ "learning_rate": 5.7358796413099635e-05,
709
+ "loss": 2.0266,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 1.5423173317910162,
714
+ "grad_norm": 0.4229032099246979,
715
+ "learning_rate": 5.646677288761132e-05,
716
+ "loss": 2.2371,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 1.5423173317910162,
721
+ "eval_loss": 2.0582103729248047,
722
+ "eval_runtime": 154.9137,
723
+ "eval_samples_per_second": 3.725,
724
+ "eval_steps_per_second": 3.725,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 1.5577405051089261,
729
+ "grad_norm": 0.565112829208374,
730
+ "learning_rate": 5.5572650989317874e-05,
731
+ "loss": 2.1034,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 1.5731636784268364,
736
+ "grad_norm": 0.5104465484619141,
737
+ "learning_rate": 5.467672084764066e-05,
738
+ "loss": 2.1003,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 1.5885868517447466,
743
+ "grad_norm": 0.49492645263671875,
744
+ "learning_rate": 5.3779273178749425e-05,
745
+ "loss": 1.9111,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 1.6040100250626566,
750
+ "grad_norm": 0.3711722493171692,
751
+ "learning_rate": 5.288059919122922e-05,
752
+ "loss": 1.8673,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 1.6194331983805668,
757
+ "grad_norm": 0.44108933210372925,
758
+ "learning_rate": 5.1980990491587465e-05,
759
+ "loss": 2.1027,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 1.634856371698477,
764
+ "grad_norm": 0.6337707042694092,
765
+ "learning_rate": 5.108073898963194e-05,
766
+ "loss": 2.1683,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 1.650279545016387,
771
+ "grad_norm": 0.4429640471935272,
772
+ "learning_rate": 5.018013680375035e-05,
773
+ "loss": 1.9772,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 1.6657027183342974,
778
+ "grad_norm": 0.4720529019832611,
779
+ "learning_rate": 4.927947616612215e-05,
780
+ "loss": 2.008,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 1.6811258916522074,
785
+ "grad_norm": 0.44429996609687805,
786
+ "learning_rate": 4.8379049327893535e-05,
787
+ "loss": 2.2032,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 1.6965490649701176,
792
+ "grad_norm": 0.39796188473701477,
793
+ "learning_rate": 4.7479148464346276e-05,
794
+ "loss": 2.0932,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 1.7119722382880278,
799
+ "grad_norm": 0.3974200189113617,
800
+ "learning_rate": 4.658006558009099e-05,
801
+ "loss": 2.0747,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 1.7273954116059378,
806
+ "grad_norm": 0.37543272972106934,
807
+ "learning_rate": 4.568209241431614e-05,
808
+ "loss": 2.3118,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 1.7428185849238482,
813
+ "grad_norm": 0.5304645299911499,
814
+ "learning_rate": 4.478552034612277e-05,
815
+ "loss": 2.047,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 1.7582417582417582,
820
+ "grad_norm": 0.43397507071495056,
821
+ "learning_rate": 4.3890640299976345e-05,
822
+ "loss": 2.1676,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 1.7736649315596684,
827
+ "grad_norm": 0.4973481297492981,
828
+ "learning_rate": 4.299774265130609e-05,
829
+ "loss": 2.023,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 1.7890881048775786,
834
+ "grad_norm": 0.4363488554954529,
835
+ "learning_rate": 4.21071171322823e-05,
836
+ "loss": 2.0945,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 1.8045112781954886,
841
+ "grad_norm": 0.4055328369140625,
842
+ "learning_rate": 4.121905273780263e-05,
843
+ "loss": 2.2178,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 1.8199344515133988,
848
+ "grad_norm": 0.49545660614967346,
849
+ "learning_rate": 4.033383763171738e-05,
850
+ "loss": 1.9995,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 1.835357624831309,
855
+ "grad_norm": 0.37422749400138855,
856
+ "learning_rate": 3.9451759053324724e-05,
857
+ "loss": 2.1181,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 1.8507807981492193,
862
+ "grad_norm": 0.37200838327407837,
863
+ "learning_rate": 3.8573103224165555e-05,
864
+ "loss": 1.917,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 1.8662039714671295,
869
+ "grad_norm": 0.4319084584712982,
870
+ "learning_rate": 3.7698155255149005e-05,
871
+ "loss": 2.165,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 1.8816271447850395,
876
+ "grad_norm": 0.3685438930988312,
877
+ "learning_rate": 3.682719905403804e-05,
878
+ "loss": 2.0359,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 1.8970503181029497,
883
+ "grad_norm": 0.6587750315666199,
884
+ "learning_rate": 3.596051723332554e-05,
885
+ "loss": 2.0902,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 1.9124734914208599,
890
+ "grad_norm": 0.5964400768280029,
891
+ "learning_rate": 3.5098391018530816e-05,
892
+ "loss": 1.9265,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 1.9278966647387699,
897
+ "grad_norm": 0.5480875372886658,
898
+ "learning_rate": 3.42411001569459e-05,
899
+ "loss": 1.9755,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 1.9433198380566803,
904
+ "grad_norm": 0.4315269887447357,
905
+ "learning_rate": 3.338892282686179e-05,
906
+ "loss": 2.0685,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 1.9587430113745903,
911
+ "grad_norm": 0.56096351146698,
912
+ "learning_rate": 3.254213554730357e-05,
913
+ "loss": 2.1827,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 1.9741661846925005,
918
+ "grad_norm": 0.5271403193473816,
919
+ "learning_rate": 3.170101308830421e-05,
920
+ "loss": 2.1641,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 1.9895893580104107,
925
+ "grad_norm": 0.4906688332557678,
926
+ "learning_rate": 3.086582838174551e-05,
927
+ "loss": 2.085,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.0050125313283207,
932
+ "grad_norm": 0.38613101840019226,
933
+ "learning_rate": 3.0036852432795925e-05,
934
+ "loss": 2.0016,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.020435704646231,
939
+ "grad_norm": 0.5777701735496521,
940
+ "learning_rate": 2.9214354231973207e-05,
941
+ "loss": 1.7941,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.035858877964141,
946
+ "grad_norm": 0.48649054765701294,
947
+ "learning_rate": 2.839860066786103e-05,
948
+ "loss": 2.1192,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 2.051282051282051,
953
+ "grad_norm": 0.4648803472518921,
954
+ "learning_rate": 2.7589856440507522e-05,
955
+ "loss": 2.0756,
956
+ "step": 1330
957
+ },
958
+ {
959
+ "epoch": 2.0667052245999615,
960
+ "grad_norm": 0.5168564319610596,
961
+ "learning_rate": 2.6788383975533994e-05,
962
+ "loss": 2.2201,
963
+ "step": 1340
964
+ },
965
+ {
966
+ "epoch": 2.0821283979178715,
967
+ "grad_norm": 0.377386212348938,
968
+ "learning_rate": 2.5994443338981732e-05,
969
+ "loss": 2.0903,
970
+ "step": 1350
971
+ },
972
+ {
973
+ "epoch": 2.097551571235782,
974
+ "grad_norm": 0.5663664937019348,
975
+ "learning_rate": 2.5208292152924262e-05,
976
+ "loss": 2.0757,
977
+ "step": 1360
978
+ },
979
+ {
980
+ "epoch": 2.112974744553692,
981
+ "grad_norm": 0.5003238320350647,
982
+ "learning_rate": 2.4430185511872944e-05,
983
+ "loss": 2.1352,
984
+ "step": 1370
985
+ },
986
+ {
987
+ "epoch": 2.128397917871602,
988
+ "grad_norm": 0.5036555528640747,
989
+ "learning_rate": 2.3660375900002362e-05,
990
+ "loss": 1.9787,
991
+ "step": 1380
992
+ },
993
+ {
994
+ "epoch": 2.1438210911895124,
995
+ "grad_norm": 0.4774503707885742,
996
+ "learning_rate": 2.2899113109223114e-05,
997
+ "loss": 1.9539,
998
+ "step": 1390
999
+ },
1000
+ {
1001
+ "epoch": 2.1592442645074224,
1002
+ "grad_norm": 0.48530662059783936,
1003
+ "learning_rate": 2.2146644158127827e-05,
1004
+ "loss": 2.179,
1005
+ "step": 1400
1006
+ },
1007
+ {
1008
+ "epoch": 2.174667437825333,
1009
+ "grad_norm": 0.5162932276725769,
1010
+ "learning_rate": 2.140321321183734e-05,
1011
+ "loss": 2.278,
1012
+ "step": 1410
1013
+ },
1014
+ {
1015
+ "epoch": 2.1900906111432428,
1016
+ "grad_norm": 0.4148849546909332,
1017
+ "learning_rate": 2.0669061502772776e-05,
1018
+ "loss": 2.1949,
1019
+ "step": 1420
1020
+ },
1021
+ {
1022
+ "epoch": 2.2055137844611528,
1023
+ "grad_norm": 0.4029206931591034,
1024
+ "learning_rate": 1.9944427252378978e-05,
1025
+ "loss": 2.1155,
1026
+ "step": 1430
1027
+ },
1028
+ {
1029
+ "epoch": 2.220936957779063,
1030
+ "grad_norm": 0.5142106413841248,
1031
+ "learning_rate": 1.9229545593825366e-05,
1032
+ "loss": 2.0562,
1033
+ "step": 1440
1034
+ },
1035
+ {
1036
+ "epoch": 2.236360131096973,
1037
+ "grad_norm": 0.4345252513885498,
1038
+ "learning_rate": 1.8524648495708512e-05,
1039
+ "loss": 2.0335,
1040
+ "step": 1450
1041
+ },
1042
+ {
1043
+ "epoch": 2.251783304414883,
1044
+ "grad_norm": 0.42935454845428467,
1045
+ "learning_rate": 1.7829964686781793e-05,
1046
+ "loss": 2.2724,
1047
+ "step": 1460
1048
+ },
1049
+ {
1050
+ "epoch": 2.2672064777327936,
1051
+ "grad_norm": 0.5543069243431091,
1052
+ "learning_rate": 1.7145719581736336e-05,
1053
+ "loss": 2.0677,
1054
+ "step": 1470
1055
+ },
1056
+ {
1057
+ "epoch": 2.2826296510507036,
1058
+ "grad_norm": 0.4770039916038513,
1059
+ "learning_rate": 1.6472135208057126e-05,
1060
+ "loss": 2.3684,
1061
+ "step": 1480
1062
+ },
1063
+ {
1064
+ "epoch": 2.298052824368614,
1065
+ "grad_norm": 0.5183836221694946,
1066
+ "learning_rate": 1.5809430133978508e-05,
1067
+ "loss": 2.0769,
1068
+ "step": 1490
1069
+ },
1070
+ {
1071
+ "epoch": 2.313475997686524,
1072
+ "grad_norm": 0.4000667929649353,
1073
+ "learning_rate": 1.5157819397561862e-05,
1074
+ "loss": 2.0174,
1075
+ "step": 1500
1076
+ },
1077
+ {
1078
+ "epoch": 2.313475997686524,
1079
+ "eval_loss": 2.042402505874634,
1080
+ "eval_runtime": 154.606,
1081
+ "eval_samples_per_second": 3.732,
1082
+ "eval_steps_per_second": 3.732,
1083
+ "step": 1500
1084
+ },
1085
+ {
1086
+ "epoch": 2.328899171004434,
1087
+ "grad_norm": 0.5184345841407776,
1088
+ "learning_rate": 1.4517514436918994e-05,
1089
+ "loss": 2.041,
1090
+ "step": 1510
1091
+ },
1092
+ {
1093
+ "epoch": 2.3443223443223444,
1094
+ "grad_norm": 0.5698699355125427,
1095
+ "learning_rate": 1.3888723021603529e-05,
1096
+ "loss": 2.026,
1097
+ "step": 1520
1098
+ },
1099
+ {
1100
+ "epoch": 2.3597455176402544,
1101
+ "grad_norm": 0.5794140100479126,
1102
+ "learning_rate": 1.327164918519282e-05,
1103
+ "loss": 2.1522,
1104
+ "step": 1530
1105
+ },
1106
+ {
1107
+ "epoch": 2.375168690958165,
1108
+ "grad_norm": 0.3988121747970581,
1109
+ "learning_rate": 1.2666493159081943e-05,
1110
+ "loss": 2.1803,
1111
+ "step": 1540
1112
+ },
1113
+ {
1114
+ "epoch": 2.390591864276075,
1115
+ "grad_norm": 0.565253496170044,
1116
+ "learning_rate": 1.207345130751164e-05,
1117
+ "loss": 2.0834,
1118
+ "step": 1550
1119
+ },
1120
+ {
1121
+ "epoch": 2.406015037593985,
1122
+ "grad_norm": 0.5080721378326416,
1123
+ "learning_rate": 1.1492716063850973e-05,
1124
+ "loss": 2.014,
1125
+ "step": 1560
1126
+ },
1127
+ {
1128
+ "epoch": 2.4214382109118953,
1129
+ "grad_norm": 0.40610289573669434,
1130
+ "learning_rate": 1.0924475868155493e-05,
1131
+ "loss": 2.0482,
1132
+ "step": 1570
1133
+ },
1134
+ {
1135
+ "epoch": 2.4368613842298052,
1136
+ "grad_norm": 0.48359382152557373,
1137
+ "learning_rate": 1.0368915106021255e-05,
1138
+ "loss": 1.9128,
1139
+ "step": 1580
1140
+ },
1141
+ {
1142
+ "epoch": 2.4522845575477152,
1143
+ "grad_norm": 0.6242773532867432,
1144
+ "learning_rate": 9.826214048754367e-06,
1145
+ "loss": 1.9792,
1146
+ "step": 1590
1147
+ },
1148
+ {
1149
+ "epoch": 2.4677077308656257,
1150
+ "grad_norm": 0.5390336513519287,
1151
+ "learning_rate": 9.29654879487566e-06,
1152
+ "loss": 1.9065,
1153
+ "step": 1600
1154
+ },
1155
+ {
1156
+ "epoch": 2.4831309041835357,
1157
+ "grad_norm": 0.49583902955055237,
1158
+ "learning_rate": 8.780091212979207e-06,
1159
+ "loss": 2.1249,
1160
+ "step": 1610
1161
+ },
1162
+ {
1163
+ "epoch": 2.498554077501446,
1164
+ "grad_norm": 0.5197187662124634,
1165
+ "learning_rate": 8.277008885963594e-06,
1166
+ "loss": 1.8338,
1167
+ "step": 1620
1168
+ },
1169
+ {
1170
+ "epoch": 2.513977250819356,
1171
+ "grad_norm": 0.5529533624649048,
1172
+ "learning_rate": 7.787465056653653e-06,
1173
+ "loss": 1.9179,
1174
+ "step": 1630
1175
+ },
1176
+ {
1177
+ "epoch": 2.5294004241372665,
1178
+ "grad_norm": 0.6007400155067444,
1179
+ "learning_rate": 7.31161857483057e-06,
1180
+ "loss": 2.0279,
1181
+ "step": 1640
1182
+ },
1183
+ {
1184
+ "epoch": 2.5448235974551765,
1185
+ "grad_norm": 0.5567540526390076,
1186
+ "learning_rate": 6.849623845687547e-06,
1187
+ "loss": 2.0689,
1188
+ "step": 1650
1189
+ },
1190
+ {
1191
+ "epoch": 2.5602467707730865,
1192
+ "grad_norm": 0.614508330821991,
1193
+ "learning_rate": 6.401630779727452e-06,
1194
+ "loss": 2.1043,
1195
+ "step": 1660
1196
+ },
1197
+ {
1198
+ "epoch": 2.575669944090997,
1199
+ "grad_norm": 0.4546593427658081,
1200
+ "learning_rate": 5.967784744119203e-06,
1201
+ "loss": 2.0116,
1202
+ "step": 1670
1203
+ },
1204
+ {
1205
+ "epoch": 2.591093117408907,
1206
+ "grad_norm": 0.35855767130851746,
1207
+ "learning_rate": 5.548226515528132e-06,
1208
+ "loss": 1.8218,
1209
+ "step": 1680
1210
+ },
1211
+ {
1212
+ "epoch": 2.606516290726817,
1213
+ "grad_norm": 0.5323745012283325,
1214
+ "learning_rate": 5.143092234436126e-06,
1215
+ "loss": 2.0448,
1216
+ "step": 1690
1217
+ },
1218
+ {
1219
+ "epoch": 2.6219394640447273,
1220
+ "grad_norm": 0.5990585684776306,
1221
+ "learning_rate": 4.752513360965949e-06,
1222
+ "loss": 1.9742,
1223
+ "step": 1700
1224
+ },
1225
+ {
1226
+ "epoch": 2.6373626373626373,
1227
+ "grad_norm": 0.5095666646957397,
1228
+ "learning_rate": 4.376616632224451e-06,
1229
+ "loss": 2.1038,
1230
+ "step": 1710
1231
+ },
1232
+ {
1233
+ "epoch": 2.6527858106805473,
1234
+ "grad_norm": 0.4896791875362396,
1235
+ "learning_rate": 4.015524021178196e-06,
1236
+ "loss": 2.0926,
1237
+ "step": 1720
1238
+ },
1239
+ {
1240
+ "epoch": 2.6682089839984577,
1241
+ "grad_norm": 0.5555339455604553,
1242
+ "learning_rate": 3.6693526970749957e-06,
1243
+ "loss": 2.1485,
1244
+ "step": 1730
1245
+ },
1246
+ {
1247
+ "epoch": 2.6836321573163677,
1248
+ "grad_norm": 0.5582926273345947,
1249
+ "learning_rate": 3.3382149874242817e-06,
1250
+ "loss": 2.1413,
1251
+ "step": 1740
1252
+ },
1253
+ {
1254
+ "epoch": 2.699055330634278,
1255
+ "grad_norm": 0.48462751507759094,
1256
+ "learning_rate": 3.0222183415484216e-06,
1257
+ "loss": 2.0053,
1258
+ "step": 1750
1259
+ },
1260
+ {
1261
+ "epoch": 2.714478503952188,
1262
+ "grad_norm": 0.5062531232833862,
1263
+ "learning_rate": 2.721465295716996e-06,
1264
+ "loss": 2.1571,
1265
+ "step": 1760
1266
+ },
1267
+ {
1268
+ "epoch": 2.7299016772700986,
1269
+ "grad_norm": 0.464996337890625,
1270
+ "learning_rate": 2.436053439875319e-06,
1271
+ "loss": 2.2241,
1272
+ "step": 1770
1273
+ },
1274
+ {
1275
+ "epoch": 2.7453248505880086,
1276
+ "grad_norm": 0.41364607214927673,
1277
+ "learning_rate": 2.1660753859779227e-06,
1278
+ "loss": 1.8589,
1279
+ "step": 1780
1280
+ },
1281
+ {
1282
+ "epoch": 2.7607480239059186,
1283
+ "grad_norm": 0.5476650595664978,
1284
+ "learning_rate": 1.911618737937304e-06,
1285
+ "loss": 2.2024,
1286
+ "step": 1790
1287
+ },
1288
+ {
1289
+ "epoch": 2.776171197223829,
1290
+ "grad_norm": 0.4496407210826874,
1291
+ "learning_rate": 1.6727660631977893e-06,
1292
+ "loss": 2.219,
1293
+ "step": 1800
1294
+ },
1295
+ {
1296
+ "epoch": 2.791594370541739,
1297
+ "grad_norm": 0.46782755851745605,
1298
+ "learning_rate": 1.4495948659435932e-06,
1299
+ "loss": 1.9196,
1300
+ "step": 1810
1301
+ },
1302
+ {
1303
+ "epoch": 2.807017543859649,
1304
+ "grad_norm": 0.5227280259132385,
1305
+ "learning_rate": 1.24217756194982e-06,
1306
+ "loss": 1.9901,
1307
+ "step": 1820
1308
+ },
1309
+ {
1310
+ "epoch": 2.8224407171775594,
1311
+ "grad_norm": 0.4377809464931488,
1312
+ "learning_rate": 1.0505814550846704e-06,
1313
+ "loss": 2.1759,
1314
+ "step": 1830
1315
+ },
1316
+ {
1317
+ "epoch": 2.8378638904954694,
1318
+ "grad_norm": 0.5370849370956421,
1319
+ "learning_rate": 8.748687154702673e-07,
1320
+ "loss": 2.0904,
1321
+ "step": 1840
1322
+ },
1323
+ {
1324
+ "epoch": 2.8532870638133794,
1325
+ "grad_norm": 0.5060635805130005,
1326
+ "learning_rate": 7.150963593094029e-07,
1327
+ "loss": 1.9761,
1328
+ "step": 1850
1329
+ },
1330
+ {
1331
+ "epoch": 2.86871023713129,
1332
+ "grad_norm": 0.49891987442970276,
1333
+ "learning_rate": 5.713162303845887e-07,
1334
+ "loss": 2.0288,
1335
+ "step": 1860
1336
+ },
1337
+ {
1338
+ "epoch": 2.8841334104492,
1339
+ "grad_norm": 0.5800537467002869,
1340
+ "learning_rate": 4.4357498323553405e-07,
1341
+ "loss": 1.9074,
1342
+ "step": 1870
1343
+ },
1344
+ {
1345
+ "epoch": 2.8995565837671102,
1346
+ "grad_norm": 0.5344066023826599,
1347
+ "learning_rate": 3.3191406802041693e-07,
1348
+ "loss": 1.8958,
1349
+ "step": 1880
1350
+ },
1351
+ {
1352
+ "epoch": 2.91497975708502,
1353
+ "grad_norm": 0.45593753457069397,
1354
+ "learning_rate": 2.3636971706592627e-07,
1355
+ "loss": 2.0741,
1356
+ "step": 1890
1357
+ },
1358
+ {
1359
+ "epoch": 2.9304029304029307,
1360
+ "grad_norm": 0.5734761953353882,
1361
+ "learning_rate": 1.5697293311039974e-07,
1362
+ "loss": 2.0763,
1363
+ "step": 1900
1364
+ },
1365
+ {
1366
+ "epoch": 2.9458261037208406,
1367
+ "grad_norm": 0.5509295463562012,
1368
+ "learning_rate": 9.374947924388755e-08,
1369
+ "loss": 1.9926,
1370
+ "step": 1910
1371
+ },
1372
+ {
1373
+ "epoch": 2.9612492770387506,
1374
+ "grad_norm": 0.4733286499977112,
1375
+ "learning_rate": 4.671987054842841e-08,
1376
+ "loss": 2.3555,
1377
+ "step": 1920
1378
+ },
1379
+ {
1380
+ "epoch": 2.976672450356661,
1381
+ "grad_norm": 0.6256054043769836,
1382
+ "learning_rate": 1.589936744116893e-08,
1383
+ "loss": 1.9345,
1384
+ "step": 1930
1385
+ },
1386
+ {
1387
+ "epoch": 2.992095623674571,
1388
+ "grad_norm": 0.5568147897720337,
1389
+ "learning_rate": 1.2979707226135062e-09,
1390
+ "loss": 2.0782,
1391
+ "step": 1940
1392
+ },
1393
+ {
1394
+ "epoch": 2.998264893001735,
1395
+ "step": 1944,
1396
+ "total_flos": 1.2799386304118784e+18,
1397
+ "train_loss": 2.14731617155389,
1398
+ "train_runtime": 14043.9874,
1399
+ "train_samples_per_second": 1.108,
1400
+ "train_steps_per_second": 0.138
1401
+ }
1402
+ ],
1403
+ "logging_steps": 10,
1404
+ "max_steps": 1944,
1405
+ "num_input_tokens_seen": 0,
1406
+ "num_train_epochs": 3,
1407
+ "save_steps": 500,
1408
+ "total_flos": 1.2799386304118784e+18,
1409
+ "train_batch_size": 1,
1410
+ "trial_name": null,
1411
+ "trial_params": null
1412
+ }
training_eval_loss.png ADDED
training_loss.png ADDED