sedrickkeh commited on
Commit
be5ffc7
·
verified ·
1 Parent(s): bbc73d1

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: top_20_ranking_stackexchange
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # top_20_ranking_stackexchange
17
 
18
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.7927
21
 
 
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: top_20_ranking_stackexchange
 
16
 
17
  # top_20_ranking_stackexchange
18
 
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/top_20_ranking_stackexchange dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.7927
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.999498746867168,
3
+ "eval_loss": 0.7926730513572693,
4
+ "eval_runtime": 796.5551,
5
+ "eval_samples_per_second": 25.308,
6
+ "eval_steps_per_second": 0.395,
7
+ "total_flos": 3758574199111680.0,
8
+ "train_loss": 0.7796513685780625,
9
+ "train_runtime": 132137.1731,
10
+ "train_samples_per_second": 8.696,
11
+ "train_steps_per_second": 0.017
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.999498746867168,
3
+ "eval_loss": 0.7926730513572693,
4
+ "eval_runtime": 796.5551,
5
+ "eval_samples_per_second": 25.308,
6
+ "eval_steps_per_second": 0.395
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.999498746867168,
3
+ "total_flos": 3758574199111680.0,
4
+ "train_loss": 0.7796513685780625,
5
+ "train_runtime": 132137.1731,
6
+ "train_samples_per_second": 8.696,
7
+ "train_steps_per_second": 0.017
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1634 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.999498746867168,
5
+ "eval_steps": 500,
6
+ "global_step": 2244,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.013366750208855471,
13
+ "grad_norm": 2.4323846059062397,
14
+ "learning_rate": 5e-06,
15
+ "loss": 1.0521,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.026733500417710943,
20
+ "grad_norm": 1.2707159898783558,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.9449,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.040100250626566414,
27
+ "grad_norm": 0.6645760066182232,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.9205,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.053467000835421885,
34
+ "grad_norm": 0.6860381528425127,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.9062,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.06683375104427736,
41
+ "grad_norm": 0.8462056832267063,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.8941,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.08020050125313283,
48
+ "grad_norm": 0.5498617128094427,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.8866,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.0935672514619883,
55
+ "grad_norm": 0.6217303867910247,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.8719,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.10693400167084377,
62
+ "grad_norm": 0.687429978149511,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.876,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.12030075187969924,
69
+ "grad_norm": 0.7639829931940186,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.8704,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.1336675020885547,
76
+ "grad_norm": 0.5349974897408032,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.8677,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.14703425229741018,
83
+ "grad_norm": 0.6212381364086903,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.8624,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.16040100250626566,
90
+ "grad_norm": 0.5610901155787884,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.8621,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.17376775271512113,
97
+ "grad_norm": 0.6155926013296407,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.8582,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.1871345029239766,
104
+ "grad_norm": 0.6528571036607788,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.8532,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.20050125313283207,
111
+ "grad_norm": 0.5372075443842537,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.8492,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.21386800334168754,
118
+ "grad_norm": 0.7095829143035569,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.8494,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.227234753550543,
125
+ "grad_norm": 0.7745444177509586,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.8476,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.24060150375939848,
132
+ "grad_norm": 0.7586050901974903,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.8494,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.25396825396825395,
139
+ "grad_norm": 0.5964597569119979,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.8498,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.2673350041771094,
146
+ "grad_norm": 0.6293549963407589,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.8432,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.2807017543859649,
153
+ "grad_norm": 0.5524407679849426,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.8475,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.29406850459482037,
160
+ "grad_norm": 0.524350214049005,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.8431,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.30743525480367584,
167
+ "grad_norm": 0.6760002252683699,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.8386,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.3208020050125313,
174
+ "grad_norm": 0.5906902446596286,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.8349,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.3341687552213868,
181
+ "grad_norm": 0.5723926384792003,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.8361,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.34753550543024225,
188
+ "grad_norm": 0.5616096712561062,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.8368,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.3609022556390977,
195
+ "grad_norm": 0.5507735559959206,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.835,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.3742690058479532,
202
+ "grad_norm": 0.4803949597709757,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.8414,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.38763575605680867,
209
+ "grad_norm": 0.5121852118343002,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.8325,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.40100250626566414,
216
+ "grad_norm": 0.5559477754717894,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.8364,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.4143692564745196,
223
+ "grad_norm": 0.7469026400245374,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.8306,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.4277360066833751,
230
+ "grad_norm": 0.5090947427034287,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.8339,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.44110275689223055,
237
+ "grad_norm": 0.6018861983279394,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.8283,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.454469507101086,
244
+ "grad_norm": 0.5434521657719814,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.8285,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.4678362573099415,
251
+ "grad_norm": 0.5903702809830117,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.8324,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.48120300751879697,
258
+ "grad_norm": 0.6243867601355255,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.8284,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.49456975772765244,
265
+ "grad_norm": 0.6094144532555286,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.8283,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.5079365079365079,
272
+ "grad_norm": 0.5482360219270039,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.8289,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.5213032581453634,
279
+ "grad_norm": 0.5061542985510644,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.8317,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.5346700083542189,
286
+ "grad_norm": 0.6652440131533577,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.8256,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.5480367585630743,
293
+ "grad_norm": 0.5613018728699922,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.8252,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.5614035087719298,
300
+ "grad_norm": 0.7255190718604577,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.8247,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.5747702589807853,
307
+ "grad_norm": 0.6781380945175464,
308
+ "learning_rate": 5e-06,
309
+ "loss": 0.823,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.5881370091896407,
314
+ "grad_norm": 0.5530197743336887,
315
+ "learning_rate": 5e-06,
316
+ "loss": 0.8251,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.6015037593984962,
321
+ "grad_norm": 0.571851888660113,
322
+ "learning_rate": 5e-06,
323
+ "loss": 0.8232,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.6148705096073517,
328
+ "grad_norm": 0.5208791337420644,
329
+ "learning_rate": 5e-06,
330
+ "loss": 0.8235,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.6282372598162071,
335
+ "grad_norm": 0.5198842932978275,
336
+ "learning_rate": 5e-06,
337
+ "loss": 0.8238,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.6416040100250626,
342
+ "grad_norm": 0.48452315583166233,
343
+ "learning_rate": 5e-06,
344
+ "loss": 0.8221,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.6549707602339181,
349
+ "grad_norm": 0.5219240912238245,
350
+ "learning_rate": 5e-06,
351
+ "loss": 0.8168,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.6683375104427736,
356
+ "grad_norm": 0.51813285089071,
357
+ "learning_rate": 5e-06,
358
+ "loss": 0.8173,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.681704260651629,
363
+ "grad_norm": 0.49897768190410446,
364
+ "learning_rate": 5e-06,
365
+ "loss": 0.8193,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.6950710108604845,
370
+ "grad_norm": 0.546834157816808,
371
+ "learning_rate": 5e-06,
372
+ "loss": 0.8129,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.70843776106934,
377
+ "grad_norm": 0.5295360571693272,
378
+ "learning_rate": 5e-06,
379
+ "loss": 0.8194,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.7218045112781954,
384
+ "grad_norm": 0.6854942956404928,
385
+ "learning_rate": 5e-06,
386
+ "loss": 0.8193,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 0.7351712614870509,
391
+ "grad_norm": 0.6819748794747951,
392
+ "learning_rate": 5e-06,
393
+ "loss": 0.8161,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 0.7485380116959064,
398
+ "grad_norm": 0.7134808000164234,
399
+ "learning_rate": 5e-06,
400
+ "loss": 0.8166,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 0.7619047619047619,
405
+ "grad_norm": 0.6412479917820569,
406
+ "learning_rate": 5e-06,
407
+ "loss": 0.8172,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 0.7752715121136173,
412
+ "grad_norm": 0.5246142664617556,
413
+ "learning_rate": 5e-06,
414
+ "loss": 0.8145,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 0.7886382623224728,
419
+ "grad_norm": 0.588843604202556,
420
+ "learning_rate": 5e-06,
421
+ "loss": 0.82,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 0.8020050125313283,
426
+ "grad_norm": 0.5124861711768851,
427
+ "learning_rate": 5e-06,
428
+ "loss": 0.8156,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 0.8153717627401837,
433
+ "grad_norm": 0.5015203839251716,
434
+ "learning_rate": 5e-06,
435
+ "loss": 0.8191,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 0.8287385129490392,
440
+ "grad_norm": 0.6441893371422894,
441
+ "learning_rate": 5e-06,
442
+ "loss": 0.812,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 0.8421052631578947,
447
+ "grad_norm": 0.5838304398634407,
448
+ "learning_rate": 5e-06,
449
+ "loss": 0.8086,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 0.8554720133667502,
454
+ "grad_norm": 0.5107304906894905,
455
+ "learning_rate": 5e-06,
456
+ "loss": 0.8155,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 0.8688387635756056,
461
+ "grad_norm": 0.5122885155184959,
462
+ "learning_rate": 5e-06,
463
+ "loss": 0.8131,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 0.8822055137844611,
468
+ "grad_norm": 0.5985811394437027,
469
+ "learning_rate": 5e-06,
470
+ "loss": 0.8104,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 0.8955722639933166,
475
+ "grad_norm": 0.5323936368547137,
476
+ "learning_rate": 5e-06,
477
+ "loss": 0.8186,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 0.908939014202172,
482
+ "grad_norm": 0.616312309430872,
483
+ "learning_rate": 5e-06,
484
+ "loss": 0.8124,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 0.9223057644110275,
489
+ "grad_norm": 0.6593022396181776,
490
+ "learning_rate": 5e-06,
491
+ "loss": 0.8156,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 0.935672514619883,
496
+ "grad_norm": 0.5181097754729659,
497
+ "learning_rate": 5e-06,
498
+ "loss": 0.8135,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 0.9490392648287385,
503
+ "grad_norm": 0.5160202542043503,
504
+ "learning_rate": 5e-06,
505
+ "loss": 0.8108,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 0.9624060150375939,
510
+ "grad_norm": 0.5439429222609182,
511
+ "learning_rate": 5e-06,
512
+ "loss": 0.8098,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 0.9757727652464494,
517
+ "grad_norm": 0.5666778381149935,
518
+ "learning_rate": 5e-06,
519
+ "loss": 0.8064,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 0.9891395154553049,
524
+ "grad_norm": 0.5087008142559319,
525
+ "learning_rate": 5e-06,
526
+ "loss": 0.8124,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 0.9998329156223893,
531
+ "eval_loss": 0.8087860345840454,
532
+ "eval_runtime": 793.9439,
533
+ "eval_samples_per_second": 25.391,
534
+ "eval_steps_per_second": 0.397,
535
+ "step": 748
536
+ },
537
+ {
538
+ "epoch": 1.0025062656641603,
539
+ "grad_norm": 0.6885103061332264,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.8763,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.0158730158730158,
546
+ "grad_norm": 0.6156521836752095,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.7692,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.0292397660818713,
553
+ "grad_norm": 0.6134559509903806,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.7719,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.0426065162907268,
560
+ "grad_norm": 0.635583159755333,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.7724,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.0559732664995822,
567
+ "grad_norm": 0.5771840092558814,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.7724,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.0693400167084377,
574
+ "grad_norm": 0.5138399093282234,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.7671,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.0827067669172932,
581
+ "grad_norm": 0.5865180500219783,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.7741,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.0960735171261486,
588
+ "grad_norm": 0.5737059877569465,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.7735,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.1094402673350041,
595
+ "grad_norm": 0.7198057887439943,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.7715,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.1228070175438596,
602
+ "grad_norm": 0.723247678442899,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.7688,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.136173767752715,
609
+ "grad_norm": 0.5724777994659187,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.7709,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.1495405179615705,
616
+ "grad_norm": 0.6343455699124487,
617
+ "learning_rate": 5e-06,
618
+ "loss": 0.7756,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.162907268170426,
623
+ "grad_norm": 0.5975092244071976,
624
+ "learning_rate": 5e-06,
625
+ "loss": 0.7762,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.1762740183792815,
630
+ "grad_norm": 0.5550810138685736,
631
+ "learning_rate": 5e-06,
632
+ "loss": 0.7713,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.189640768588137,
637
+ "grad_norm": 0.6031833100946619,
638
+ "learning_rate": 5e-06,
639
+ "loss": 0.7717,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 1.2030075187969924,
644
+ "grad_norm": 0.5674692784021945,
645
+ "learning_rate": 5e-06,
646
+ "loss": 0.7714,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 1.2163742690058479,
651
+ "grad_norm": 0.6831373781930358,
652
+ "learning_rate": 5e-06,
653
+ "loss": 0.7727,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 1.2297410192147034,
658
+ "grad_norm": 0.517398562451772,
659
+ "learning_rate": 5e-06,
660
+ "loss": 0.7715,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 1.2431077694235588,
665
+ "grad_norm": 0.5689793551691444,
666
+ "learning_rate": 5e-06,
667
+ "loss": 0.7682,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 1.2564745196324143,
672
+ "grad_norm": 0.6979997189308218,
673
+ "learning_rate": 5e-06,
674
+ "loss": 0.7753,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 1.2698412698412698,
679
+ "grad_norm": 0.5431703707142987,
680
+ "learning_rate": 5e-06,
681
+ "loss": 0.7726,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 1.2832080200501252,
686
+ "grad_norm": 0.5341233588300426,
687
+ "learning_rate": 5e-06,
688
+ "loss": 0.7721,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 1.2965747702589807,
693
+ "grad_norm": 0.5621957425809071,
694
+ "learning_rate": 5e-06,
695
+ "loss": 0.7702,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 1.3099415204678362,
700
+ "grad_norm": 0.6187116295591158,
701
+ "learning_rate": 5e-06,
702
+ "loss": 0.7755,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 1.3233082706766917,
707
+ "grad_norm": 0.6251656247161459,
708
+ "learning_rate": 5e-06,
709
+ "loss": 0.7742,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 1.3366750208855471,
714
+ "grad_norm": 0.6092934361550684,
715
+ "learning_rate": 5e-06,
716
+ "loss": 0.7732,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 1.3500417710944026,
721
+ "grad_norm": 0.8086073910477094,
722
+ "learning_rate": 5e-06,
723
+ "loss": 0.7663,
724
+ "step": 1010
725
+ },
726
+ {
727
+ "epoch": 1.363408521303258,
728
+ "grad_norm": 0.6337909009600926,
729
+ "learning_rate": 5e-06,
730
+ "loss": 0.7698,
731
+ "step": 1020
732
+ },
733
+ {
734
+ "epoch": 1.3767752715121135,
735
+ "grad_norm": 0.6156017975821142,
736
+ "learning_rate": 5e-06,
737
+ "loss": 0.7687,
738
+ "step": 1030
739
+ },
740
+ {
741
+ "epoch": 1.390142021720969,
742
+ "grad_norm": 0.4791494199069362,
743
+ "learning_rate": 5e-06,
744
+ "loss": 0.7707,
745
+ "step": 1040
746
+ },
747
+ {
748
+ "epoch": 1.4035087719298245,
749
+ "grad_norm": 0.5102907384647386,
750
+ "learning_rate": 5e-06,
751
+ "loss": 0.7698,
752
+ "step": 1050
753
+ },
754
+ {
755
+ "epoch": 1.41687552213868,
756
+ "grad_norm": 0.60763231448239,
757
+ "learning_rate": 5e-06,
758
+ "loss": 0.7722,
759
+ "step": 1060
760
+ },
761
+ {
762
+ "epoch": 1.4302422723475354,
763
+ "grad_norm": 0.5538961425736992,
764
+ "learning_rate": 5e-06,
765
+ "loss": 0.7769,
766
+ "step": 1070
767
+ },
768
+ {
769
+ "epoch": 1.443609022556391,
770
+ "grad_norm": 0.511489662319519,
771
+ "learning_rate": 5e-06,
772
+ "loss": 0.7709,
773
+ "step": 1080
774
+ },
775
+ {
776
+ "epoch": 1.4569757727652464,
777
+ "grad_norm": 0.5006381424370965,
778
+ "learning_rate": 5e-06,
779
+ "loss": 0.7652,
780
+ "step": 1090
781
+ },
782
+ {
783
+ "epoch": 1.4703425229741018,
784
+ "grad_norm": 0.6446877306415851,
785
+ "learning_rate": 5e-06,
786
+ "loss": 0.7668,
787
+ "step": 1100
788
+ },
789
+ {
790
+ "epoch": 1.4837092731829573,
791
+ "grad_norm": 0.6472792025046472,
792
+ "learning_rate": 5e-06,
793
+ "loss": 0.7748,
794
+ "step": 1110
795
+ },
796
+ {
797
+ "epoch": 1.4970760233918128,
798
+ "grad_norm": 0.5297094594069526,
799
+ "learning_rate": 5e-06,
800
+ "loss": 0.7716,
801
+ "step": 1120
802
+ },
803
+ {
804
+ "epoch": 1.5104427736006683,
805
+ "grad_norm": 0.5172754876638852,
806
+ "learning_rate": 5e-06,
807
+ "loss": 0.7693,
808
+ "step": 1130
809
+ },
810
+ {
811
+ "epoch": 1.5238095238095237,
812
+ "grad_norm": 0.5499645842959932,
813
+ "learning_rate": 5e-06,
814
+ "loss": 0.7663,
815
+ "step": 1140
816
+ },
817
+ {
818
+ "epoch": 1.5371762740183792,
819
+ "grad_norm": 0.5115786493746641,
820
+ "learning_rate": 5e-06,
821
+ "loss": 0.7707,
822
+ "step": 1150
823
+ },
824
+ {
825
+ "epoch": 1.5505430242272347,
826
+ "grad_norm": 0.5733666230248589,
827
+ "learning_rate": 5e-06,
828
+ "loss": 0.7708,
829
+ "step": 1160
830
+ },
831
+ {
832
+ "epoch": 1.5639097744360901,
833
+ "grad_norm": 0.4914243878129098,
834
+ "learning_rate": 5e-06,
835
+ "loss": 0.769,
836
+ "step": 1170
837
+ },
838
+ {
839
+ "epoch": 1.5772765246449456,
840
+ "grad_norm": 0.5986514689445189,
841
+ "learning_rate": 5e-06,
842
+ "loss": 0.7722,
843
+ "step": 1180
844
+ },
845
+ {
846
+ "epoch": 1.590643274853801,
847
+ "grad_norm": 0.49301214049058534,
848
+ "learning_rate": 5e-06,
849
+ "loss": 0.7709,
850
+ "step": 1190
851
+ },
852
+ {
853
+ "epoch": 1.6040100250626566,
854
+ "grad_norm": 0.49122462674305145,
855
+ "learning_rate": 5e-06,
856
+ "loss": 0.7684,
857
+ "step": 1200
858
+ },
859
+ {
860
+ "epoch": 1.617376775271512,
861
+ "grad_norm": 0.5231320343494373,
862
+ "learning_rate": 5e-06,
863
+ "loss": 0.773,
864
+ "step": 1210
865
+ },
866
+ {
867
+ "epoch": 1.6307435254803675,
868
+ "grad_norm": 0.5974519524827527,
869
+ "learning_rate": 5e-06,
870
+ "loss": 0.7703,
871
+ "step": 1220
872
+ },
873
+ {
874
+ "epoch": 1.644110275689223,
875
+ "grad_norm": 0.49755848059450075,
876
+ "learning_rate": 5e-06,
877
+ "loss": 0.7684,
878
+ "step": 1230
879
+ },
880
+ {
881
+ "epoch": 1.6574770258980784,
882
+ "grad_norm": 0.49980350150699104,
883
+ "learning_rate": 5e-06,
884
+ "loss": 0.7648,
885
+ "step": 1240
886
+ },
887
+ {
888
+ "epoch": 1.670843776106934,
889
+ "grad_norm": 0.660197673406872,
890
+ "learning_rate": 5e-06,
891
+ "loss": 0.7663,
892
+ "step": 1250
893
+ },
894
+ {
895
+ "epoch": 1.6842105263157894,
896
+ "grad_norm": 0.501447743813946,
897
+ "learning_rate": 5e-06,
898
+ "loss": 0.7687,
899
+ "step": 1260
900
+ },
901
+ {
902
+ "epoch": 1.6975772765246449,
903
+ "grad_norm": 0.47339053427865196,
904
+ "learning_rate": 5e-06,
905
+ "loss": 0.7677,
906
+ "step": 1270
907
+ },
908
+ {
909
+ "epoch": 1.7109440267335003,
910
+ "grad_norm": 0.4776630843112484,
911
+ "learning_rate": 5e-06,
912
+ "loss": 0.7705,
913
+ "step": 1280
914
+ },
915
+ {
916
+ "epoch": 1.7243107769423558,
917
+ "grad_norm": 0.5805611285838953,
918
+ "learning_rate": 5e-06,
919
+ "loss": 0.7664,
920
+ "step": 1290
921
+ },
922
+ {
923
+ "epoch": 1.7376775271512113,
924
+ "grad_norm": 0.5589747352729452,
925
+ "learning_rate": 5e-06,
926
+ "loss": 0.7643,
927
+ "step": 1300
928
+ },
929
+ {
930
+ "epoch": 1.7510442773600667,
931
+ "grad_norm": 0.5862892637271495,
932
+ "learning_rate": 5e-06,
933
+ "loss": 0.767,
934
+ "step": 1310
935
+ },
936
+ {
937
+ "epoch": 1.7644110275689222,
938
+ "grad_norm": 0.6267084370944045,
939
+ "learning_rate": 5e-06,
940
+ "loss": 0.7701,
941
+ "step": 1320
942
+ },
943
+ {
944
+ "epoch": 1.7777777777777777,
945
+ "grad_norm": 0.5590629149887701,
946
+ "learning_rate": 5e-06,
947
+ "loss": 0.7725,
948
+ "step": 1330
949
+ },
950
+ {
951
+ "epoch": 1.7911445279866332,
952
+ "grad_norm": 0.589200505231269,
953
+ "learning_rate": 5e-06,
954
+ "loss": 0.768,
955
+ "step": 1340
956
+ },
957
+ {
958
+ "epoch": 1.8045112781954886,
959
+ "grad_norm": 0.4948446583957624,
960
+ "learning_rate": 5e-06,
961
+ "loss": 0.7685,
962
+ "step": 1350
963
+ },
964
+ {
965
+ "epoch": 1.817878028404344,
966
+ "grad_norm": 0.471229575382462,
967
+ "learning_rate": 5e-06,
968
+ "loss": 0.7685,
969
+ "step": 1360
970
+ },
971
+ {
972
+ "epoch": 1.8312447786131996,
973
+ "grad_norm": 0.5347363048336566,
974
+ "learning_rate": 5e-06,
975
+ "loss": 0.7668,
976
+ "step": 1370
977
+ },
978
+ {
979
+ "epoch": 1.844611528822055,
980
+ "grad_norm": 0.6085798758140744,
981
+ "learning_rate": 5e-06,
982
+ "loss": 0.7685,
983
+ "step": 1380
984
+ },
985
+ {
986
+ "epoch": 1.8579782790309105,
987
+ "grad_norm": 0.49237779847072155,
988
+ "learning_rate": 5e-06,
989
+ "loss": 0.766,
990
+ "step": 1390
991
+ },
992
+ {
993
+ "epoch": 1.871345029239766,
994
+ "grad_norm": 0.5429938063483495,
995
+ "learning_rate": 5e-06,
996
+ "loss": 0.7675,
997
+ "step": 1400
998
+ },
999
+ {
1000
+ "epoch": 1.8847117794486214,
1001
+ "grad_norm": 0.5315522378087794,
1002
+ "learning_rate": 5e-06,
1003
+ "loss": 0.7651,
1004
+ "step": 1410
1005
+ },
1006
+ {
1007
+ "epoch": 1.898078529657477,
1008
+ "grad_norm": 0.5774851920268103,
1009
+ "learning_rate": 5e-06,
1010
+ "loss": 0.7683,
1011
+ "step": 1420
1012
+ },
1013
+ {
1014
+ "epoch": 1.9114452798663324,
1015
+ "grad_norm": 0.4774206459938876,
1016
+ "learning_rate": 5e-06,
1017
+ "loss": 0.7651,
1018
+ "step": 1430
1019
+ },
1020
+ {
1021
+ "epoch": 1.9248120300751879,
1022
+ "grad_norm": 0.48893280928600313,
1023
+ "learning_rate": 5e-06,
1024
+ "loss": 0.7664,
1025
+ "step": 1440
1026
+ },
1027
+ {
1028
+ "epoch": 1.9381787802840433,
1029
+ "grad_norm": 0.47709822943051283,
1030
+ "learning_rate": 5e-06,
1031
+ "loss": 0.7667,
1032
+ "step": 1450
1033
+ },
1034
+ {
1035
+ "epoch": 1.9515455304928988,
1036
+ "grad_norm": 0.5221458173728611,
1037
+ "learning_rate": 5e-06,
1038
+ "loss": 0.7649,
1039
+ "step": 1460
1040
+ },
1041
+ {
1042
+ "epoch": 1.9649122807017543,
1043
+ "grad_norm": 0.5458985479332612,
1044
+ "learning_rate": 5e-06,
1045
+ "loss": 0.7653,
1046
+ "step": 1470
1047
+ },
1048
+ {
1049
+ "epoch": 1.9782790309106097,
1050
+ "grad_norm": 0.5449151757658263,
1051
+ "learning_rate": 5e-06,
1052
+ "loss": 0.7665,
1053
+ "step": 1480
1054
+ },
1055
+ {
1056
+ "epoch": 1.9916457811194652,
1057
+ "grad_norm": 0.5792068417255367,
1058
+ "learning_rate": 5e-06,
1059
+ "loss": 0.7674,
1060
+ "step": 1490
1061
+ },
1062
+ {
1063
+ "epoch": 1.9996658312447786,
1064
+ "eval_loss": 0.7951143383979797,
1065
+ "eval_runtime": 795.386,
1066
+ "eval_samples_per_second": 25.345,
1067
+ "eval_steps_per_second": 0.396,
1068
+ "step": 1496
1069
+ },
1070
+ {
1071
+ "epoch": 2.0050125313283207,
1072
+ "grad_norm": 0.7521880602206925,
1073
+ "learning_rate": 5e-06,
1074
+ "loss": 0.8233,
1075
+ "step": 1500
1076
+ },
1077
+ {
1078
+ "epoch": 2.018379281537176,
1079
+ "grad_norm": 0.6560054074439666,
1080
+ "learning_rate": 5e-06,
1081
+ "loss": 0.7256,
1082
+ "step": 1510
1083
+ },
1084
+ {
1085
+ "epoch": 2.0317460317460316,
1086
+ "grad_norm": 0.5201512747130638,
1087
+ "learning_rate": 5e-06,
1088
+ "loss": 0.7218,
1089
+ "step": 1520
1090
+ },
1091
+ {
1092
+ "epoch": 2.045112781954887,
1093
+ "grad_norm": 0.5262590120532872,
1094
+ "learning_rate": 5e-06,
1095
+ "loss": 0.7285,
1096
+ "step": 1530
1097
+ },
1098
+ {
1099
+ "epoch": 2.0584795321637426,
1100
+ "grad_norm": 0.5393650388873087,
1101
+ "learning_rate": 5e-06,
1102
+ "loss": 0.7229,
1103
+ "step": 1540
1104
+ },
1105
+ {
1106
+ "epoch": 2.071846282372598,
1107
+ "grad_norm": 0.5105428821348765,
1108
+ "learning_rate": 5e-06,
1109
+ "loss": 0.7231,
1110
+ "step": 1550
1111
+ },
1112
+ {
1113
+ "epoch": 2.0852130325814535,
1114
+ "grad_norm": 0.6021970483052078,
1115
+ "learning_rate": 5e-06,
1116
+ "loss": 0.7239,
1117
+ "step": 1560
1118
+ },
1119
+ {
1120
+ "epoch": 2.098579782790309,
1121
+ "grad_norm": 0.5009099309313954,
1122
+ "learning_rate": 5e-06,
1123
+ "loss": 0.7226,
1124
+ "step": 1570
1125
+ },
1126
+ {
1127
+ "epoch": 2.1119465329991645,
1128
+ "grad_norm": 0.5605434690720502,
1129
+ "learning_rate": 5e-06,
1130
+ "loss": 0.7277,
1131
+ "step": 1580
1132
+ },
1133
+ {
1134
+ "epoch": 2.12531328320802,
1135
+ "grad_norm": 0.5732299598938305,
1136
+ "learning_rate": 5e-06,
1137
+ "loss": 0.7286,
1138
+ "step": 1590
1139
+ },
1140
+ {
1141
+ "epoch": 2.1386800334168754,
1142
+ "grad_norm": 0.5399334511302041,
1143
+ "learning_rate": 5e-06,
1144
+ "loss": 0.726,
1145
+ "step": 1600
1146
+ },
1147
+ {
1148
+ "epoch": 2.152046783625731,
1149
+ "grad_norm": 0.505832452848056,
1150
+ "learning_rate": 5e-06,
1151
+ "loss": 0.7304,
1152
+ "step": 1610
1153
+ },
1154
+ {
1155
+ "epoch": 2.1654135338345863,
1156
+ "grad_norm": 0.5674143618926153,
1157
+ "learning_rate": 5e-06,
1158
+ "loss": 0.7232,
1159
+ "step": 1620
1160
+ },
1161
+ {
1162
+ "epoch": 2.178780284043442,
1163
+ "grad_norm": 0.5068914103748654,
1164
+ "learning_rate": 5e-06,
1165
+ "loss": 0.7336,
1166
+ "step": 1630
1167
+ },
1168
+ {
1169
+ "epoch": 2.1921470342522973,
1170
+ "grad_norm": 0.5118320329600874,
1171
+ "learning_rate": 5e-06,
1172
+ "loss": 0.7255,
1173
+ "step": 1640
1174
+ },
1175
+ {
1176
+ "epoch": 2.2055137844611528,
1177
+ "grad_norm": 0.5156250232792499,
1178
+ "learning_rate": 5e-06,
1179
+ "loss": 0.7295,
1180
+ "step": 1650
1181
+ },
1182
+ {
1183
+ "epoch": 2.2188805346700082,
1184
+ "grad_norm": 0.6165225897496419,
1185
+ "learning_rate": 5e-06,
1186
+ "loss": 0.7274,
1187
+ "step": 1660
1188
+ },
1189
+ {
1190
+ "epoch": 2.2322472848788637,
1191
+ "grad_norm": 0.5863877720536036,
1192
+ "learning_rate": 5e-06,
1193
+ "loss": 0.7256,
1194
+ "step": 1670
1195
+ },
1196
+ {
1197
+ "epoch": 2.245614035087719,
1198
+ "grad_norm": 0.5641007704480012,
1199
+ "learning_rate": 5e-06,
1200
+ "loss": 0.7308,
1201
+ "step": 1680
1202
+ },
1203
+ {
1204
+ "epoch": 2.2589807852965746,
1205
+ "grad_norm": 0.6101312501534099,
1206
+ "learning_rate": 5e-06,
1207
+ "loss": 0.7314,
1208
+ "step": 1690
1209
+ },
1210
+ {
1211
+ "epoch": 2.27234753550543,
1212
+ "grad_norm": 0.5200998469176243,
1213
+ "learning_rate": 5e-06,
1214
+ "loss": 0.7275,
1215
+ "step": 1700
1216
+ },
1217
+ {
1218
+ "epoch": 2.2857142857142856,
1219
+ "grad_norm": 0.5398343134194046,
1220
+ "learning_rate": 5e-06,
1221
+ "loss": 0.727,
1222
+ "step": 1710
1223
+ },
1224
+ {
1225
+ "epoch": 2.299081035923141,
1226
+ "grad_norm": 0.5247712631574941,
1227
+ "learning_rate": 5e-06,
1228
+ "loss": 0.727,
1229
+ "step": 1720
1230
+ },
1231
+ {
1232
+ "epoch": 2.3124477861319965,
1233
+ "grad_norm": 0.5655985095958795,
1234
+ "learning_rate": 5e-06,
1235
+ "loss": 0.7286,
1236
+ "step": 1730
1237
+ },
1238
+ {
1239
+ "epoch": 2.325814536340852,
1240
+ "grad_norm": 0.5927409653328921,
1241
+ "learning_rate": 5e-06,
1242
+ "loss": 0.7271,
1243
+ "step": 1740
1244
+ },
1245
+ {
1246
+ "epoch": 2.3391812865497075,
1247
+ "grad_norm": 0.6148593425957483,
1248
+ "learning_rate": 5e-06,
1249
+ "loss": 0.733,
1250
+ "step": 1750
1251
+ },
1252
+ {
1253
+ "epoch": 2.352548036758563,
1254
+ "grad_norm": 0.5969831864554942,
1255
+ "learning_rate": 5e-06,
1256
+ "loss": 0.7302,
1257
+ "step": 1760
1258
+ },
1259
+ {
1260
+ "epoch": 2.3659147869674184,
1261
+ "grad_norm": 0.4985456007136878,
1262
+ "learning_rate": 5e-06,
1263
+ "loss": 0.7341,
1264
+ "step": 1770
1265
+ },
1266
+ {
1267
+ "epoch": 2.379281537176274,
1268
+ "grad_norm": 0.5005254522981937,
1269
+ "learning_rate": 5e-06,
1270
+ "loss": 0.7244,
1271
+ "step": 1780
1272
+ },
1273
+ {
1274
+ "epoch": 2.3926482873851294,
1275
+ "grad_norm": 0.5288709360617612,
1276
+ "learning_rate": 5e-06,
1277
+ "loss": 0.7312,
1278
+ "step": 1790
1279
+ },
1280
+ {
1281
+ "epoch": 2.406015037593985,
1282
+ "grad_norm": 0.5355584900475018,
1283
+ "learning_rate": 5e-06,
1284
+ "loss": 0.727,
1285
+ "step": 1800
1286
+ },
1287
+ {
1288
+ "epoch": 2.4193817878028403,
1289
+ "grad_norm": 0.5666733459714918,
1290
+ "learning_rate": 5e-06,
1291
+ "loss": 0.731,
1292
+ "step": 1810
1293
+ },
1294
+ {
1295
+ "epoch": 2.4327485380116958,
1296
+ "grad_norm": 0.5939862506331437,
1297
+ "learning_rate": 5e-06,
1298
+ "loss": 0.7292,
1299
+ "step": 1820
1300
+ },
1301
+ {
1302
+ "epoch": 2.4461152882205512,
1303
+ "grad_norm": 0.5696153125681646,
1304
+ "learning_rate": 5e-06,
1305
+ "loss": 0.7295,
1306
+ "step": 1830
1307
+ },
1308
+ {
1309
+ "epoch": 2.4594820384294067,
1310
+ "grad_norm": 0.5263801998302109,
1311
+ "learning_rate": 5e-06,
1312
+ "loss": 0.7289,
1313
+ "step": 1840
1314
+ },
1315
+ {
1316
+ "epoch": 2.472848788638262,
1317
+ "grad_norm": 0.5564137280433736,
1318
+ "learning_rate": 5e-06,
1319
+ "loss": 0.7289,
1320
+ "step": 1850
1321
+ },
1322
+ {
1323
+ "epoch": 2.4862155388471177,
1324
+ "grad_norm": 0.6117589560276474,
1325
+ "learning_rate": 5e-06,
1326
+ "loss": 0.7281,
1327
+ "step": 1860
1328
+ },
1329
+ {
1330
+ "epoch": 2.499582289055973,
1331
+ "grad_norm": 0.5556838242891475,
1332
+ "learning_rate": 5e-06,
1333
+ "loss": 0.7296,
1334
+ "step": 1870
1335
+ },
1336
+ {
1337
+ "epoch": 2.5129490392648286,
1338
+ "grad_norm": 0.4681598446789898,
1339
+ "learning_rate": 5e-06,
1340
+ "loss": 0.7296,
1341
+ "step": 1880
1342
+ },
1343
+ {
1344
+ "epoch": 2.526315789473684,
1345
+ "grad_norm": 0.5231611697501862,
1346
+ "learning_rate": 5e-06,
1347
+ "loss": 0.7303,
1348
+ "step": 1890
1349
+ },
1350
+ {
1351
+ "epoch": 2.5396825396825395,
1352
+ "grad_norm": 0.5126109088017671,
1353
+ "learning_rate": 5e-06,
1354
+ "loss": 0.7324,
1355
+ "step": 1900
1356
+ },
1357
+ {
1358
+ "epoch": 2.553049289891395,
1359
+ "grad_norm": 0.5300428577804921,
1360
+ "learning_rate": 5e-06,
1361
+ "loss": 0.7273,
1362
+ "step": 1910
1363
+ },
1364
+ {
1365
+ "epoch": 2.5664160401002505,
1366
+ "grad_norm": 0.4968055663040118,
1367
+ "learning_rate": 5e-06,
1368
+ "loss": 0.729,
1369
+ "step": 1920
1370
+ },
1371
+ {
1372
+ "epoch": 2.579782790309106,
1373
+ "grad_norm": 0.568494743059541,
1374
+ "learning_rate": 5e-06,
1375
+ "loss": 0.7269,
1376
+ "step": 1930
1377
+ },
1378
+ {
1379
+ "epoch": 2.5931495405179614,
1380
+ "grad_norm": 0.5482221484283202,
1381
+ "learning_rate": 5e-06,
1382
+ "loss": 0.7285,
1383
+ "step": 1940
1384
+ },
1385
+ {
1386
+ "epoch": 2.606516290726817,
1387
+ "grad_norm": 0.47129332867964935,
1388
+ "learning_rate": 5e-06,
1389
+ "loss": 0.7292,
1390
+ "step": 1950
1391
+ },
1392
+ {
1393
+ "epoch": 2.6198830409356724,
1394
+ "grad_norm": 0.5198836974979396,
1395
+ "learning_rate": 5e-06,
1396
+ "loss": 0.7264,
1397
+ "step": 1960
1398
+ },
1399
+ {
1400
+ "epoch": 2.633249791144528,
1401
+ "grad_norm": 0.4945939304862693,
1402
+ "learning_rate": 5e-06,
1403
+ "loss": 0.7279,
1404
+ "step": 1970
1405
+ },
1406
+ {
1407
+ "epoch": 2.6466165413533833,
1408
+ "grad_norm": 0.5751403403674279,
1409
+ "learning_rate": 5e-06,
1410
+ "loss": 0.7282,
1411
+ "step": 1980
1412
+ },
1413
+ {
1414
+ "epoch": 2.659983291562239,
1415
+ "grad_norm": 0.5611452949151137,
1416
+ "learning_rate": 5e-06,
1417
+ "loss": 0.7331,
1418
+ "step": 1990
1419
+ },
1420
+ {
1421
+ "epoch": 2.6733500417710943,
1422
+ "grad_norm": 0.6119128996618558,
1423
+ "learning_rate": 5e-06,
1424
+ "loss": 0.7296,
1425
+ "step": 2000
1426
+ },
1427
+ {
1428
+ "epoch": 2.6867167919799497,
1429
+ "grad_norm": 0.4799215562608329,
1430
+ "learning_rate": 5e-06,
1431
+ "loss": 0.7298,
1432
+ "step": 2010
1433
+ },
1434
+ {
1435
+ "epoch": 2.700083542188805,
1436
+ "grad_norm": 0.5541418078345739,
1437
+ "learning_rate": 5e-06,
1438
+ "loss": 0.7268,
1439
+ "step": 2020
1440
+ },
1441
+ {
1442
+ "epoch": 2.7134502923976607,
1443
+ "grad_norm": 0.6870311878219804,
1444
+ "learning_rate": 5e-06,
1445
+ "loss": 0.7277,
1446
+ "step": 2030
1447
+ },
1448
+ {
1449
+ "epoch": 2.726817042606516,
1450
+ "grad_norm": 0.5687894755714459,
1451
+ "learning_rate": 5e-06,
1452
+ "loss": 0.7298,
1453
+ "step": 2040
1454
+ },
1455
+ {
1456
+ "epoch": 2.7401837928153716,
1457
+ "grad_norm": 0.5330460246090263,
1458
+ "learning_rate": 5e-06,
1459
+ "loss": 0.7325,
1460
+ "step": 2050
1461
+ },
1462
+ {
1463
+ "epoch": 2.753550543024227,
1464
+ "grad_norm": 0.5427879116319339,
1465
+ "learning_rate": 5e-06,
1466
+ "loss": 0.7296,
1467
+ "step": 2060
1468
+ },
1469
+ {
1470
+ "epoch": 2.7669172932330826,
1471
+ "grad_norm": 0.6013738539276209,
1472
+ "learning_rate": 5e-06,
1473
+ "loss": 0.7281,
1474
+ "step": 2070
1475
+ },
1476
+ {
1477
+ "epoch": 2.780284043441938,
1478
+ "grad_norm": 0.6091854363964149,
1479
+ "learning_rate": 5e-06,
1480
+ "loss": 0.7294,
1481
+ "step": 2080
1482
+ },
1483
+ {
1484
+ "epoch": 2.7936507936507935,
1485
+ "grad_norm": 0.5190279913663577,
1486
+ "learning_rate": 5e-06,
1487
+ "loss": 0.7248,
1488
+ "step": 2090
1489
+ },
1490
+ {
1491
+ "epoch": 2.807017543859649,
1492
+ "grad_norm": 0.5126718278939274,
1493
+ "learning_rate": 5e-06,
1494
+ "loss": 0.7311,
1495
+ "step": 2100
1496
+ },
1497
+ {
1498
+ "epoch": 2.8203842940685044,
1499
+ "grad_norm": 0.5571607138857257,
1500
+ "learning_rate": 5e-06,
1501
+ "loss": 0.7318,
1502
+ "step": 2110
1503
+ },
1504
+ {
1505
+ "epoch": 2.83375104427736,
1506
+ "grad_norm": 0.5341175882686895,
1507
+ "learning_rate": 5e-06,
1508
+ "loss": 0.7336,
1509
+ "step": 2120
1510
+ },
1511
+ {
1512
+ "epoch": 2.8471177944862154,
1513
+ "grad_norm": 0.4817774606348232,
1514
+ "learning_rate": 5e-06,
1515
+ "loss": 0.731,
1516
+ "step": 2130
1517
+ },
1518
+ {
1519
+ "epoch": 2.860484544695071,
1520
+ "grad_norm": 0.5487220776810837,
1521
+ "learning_rate": 5e-06,
1522
+ "loss": 0.7282,
1523
+ "step": 2140
1524
+ },
1525
+ {
1526
+ "epoch": 2.8738512949039263,
1527
+ "grad_norm": 0.6342699103351254,
1528
+ "learning_rate": 5e-06,
1529
+ "loss": 0.7335,
1530
+ "step": 2150
1531
+ },
1532
+ {
1533
+ "epoch": 2.887218045112782,
1534
+ "grad_norm": 0.5078552425291176,
1535
+ "learning_rate": 5e-06,
1536
+ "loss": 0.7273,
1537
+ "step": 2160
1538
+ },
1539
+ {
1540
+ "epoch": 2.9005847953216373,
1541
+ "grad_norm": 0.4819316377635323,
1542
+ "learning_rate": 5e-06,
1543
+ "loss": 0.7332,
1544
+ "step": 2170
1545
+ },
1546
+ {
1547
+ "epoch": 2.9139515455304927,
1548
+ "grad_norm": 0.4627017239179797,
1549
+ "learning_rate": 5e-06,
1550
+ "loss": 0.7306,
1551
+ "step": 2180
1552
+ },
1553
+ {
1554
+ "epoch": 2.927318295739348,
1555
+ "grad_norm": 0.4761325291977869,
1556
+ "learning_rate": 5e-06,
1557
+ "loss": 0.7314,
1558
+ "step": 2190
1559
+ },
1560
+ {
1561
+ "epoch": 2.9406850459482037,
1562
+ "grad_norm": 0.5784029020001881,
1563
+ "learning_rate": 5e-06,
1564
+ "loss": 0.7298,
1565
+ "step": 2200
1566
+ },
1567
+ {
1568
+ "epoch": 2.954051796157059,
1569
+ "grad_norm": 0.5120822643666457,
1570
+ "learning_rate": 5e-06,
1571
+ "loss": 0.731,
1572
+ "step": 2210
1573
+ },
1574
+ {
1575
+ "epoch": 2.9674185463659146,
1576
+ "grad_norm": 0.5116915736315969,
1577
+ "learning_rate": 5e-06,
1578
+ "loss": 0.7322,
1579
+ "step": 2220
1580
+ },
1581
+ {
1582
+ "epoch": 2.98078529657477,
1583
+ "grad_norm": 0.5021133290964584,
1584
+ "learning_rate": 5e-06,
1585
+ "loss": 0.7269,
1586
+ "step": 2230
1587
+ },
1588
+ {
1589
+ "epoch": 2.9941520467836256,
1590
+ "grad_norm": 0.5317540745896701,
1591
+ "learning_rate": 5e-06,
1592
+ "loss": 0.7322,
1593
+ "step": 2240
1594
+ },
1595
+ {
1596
+ "epoch": 2.999498746867168,
1597
+ "eval_loss": 0.7926730513572693,
1598
+ "eval_runtime": 792.6639,
1599
+ "eval_samples_per_second": 25.432,
1600
+ "eval_steps_per_second": 0.397,
1601
+ "step": 2244
1602
+ },
1603
+ {
1604
+ "epoch": 2.999498746867168,
1605
+ "step": 2244,
1606
+ "total_flos": 3758574199111680.0,
1607
+ "train_loss": 0.7796513685780625,
1608
+ "train_runtime": 132137.1731,
1609
+ "train_samples_per_second": 8.696,
1610
+ "train_steps_per_second": 0.017
1611
+ }
1612
+ ],
1613
+ "logging_steps": 10,
1614
+ "max_steps": 2244,
1615
+ "num_input_tokens_seen": 0,
1616
+ "num_train_epochs": 3,
1617
+ "save_steps": 500,
1618
+ "stateful_callbacks": {
1619
+ "TrainerControl": {
1620
+ "args": {
1621
+ "should_epoch_stop": false,
1622
+ "should_evaluate": false,
1623
+ "should_log": false,
1624
+ "should_save": true,
1625
+ "should_training_stop": true
1626
+ },
1627
+ "attributes": {}
1628
+ }
1629
+ },
1630
+ "total_flos": 3758574199111680.0,
1631
+ "train_batch_size": 8,
1632
+ "trial_name": null,
1633
+ "trial_params": null
1634
+ }
training_eval_loss.png ADDED
training_loss.png ADDED