Augusto777
commited on
Commit
•
f73dbac
1
Parent(s):
8db8813
End of training
Browse files- README.md +2 -2
- all_results.json +9 -9
- eval_results.json +5 -5
- runs/Feb01_19-16-14_b2db6ba6b423/events.out.tfevents.1706815248.b2db6ba6b423.3854.1 +3 -0
- train_results.json +4 -4
- trainer_state.json +215 -215
README.md
CHANGED
@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
This model is a fine-tuned version of [MBZUAI/swiftformer-xs](https://huggingface.co/MBZUAI/swiftformer-xs) on an unknown dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
-
- Loss: 0.
|
20 |
-
- Accuracy: 0.
|
21 |
|
22 |
## Model description
|
23 |
|
|
|
16 |
|
17 |
This model is a fine-tuned version of [MBZUAI/swiftformer-xs](https://huggingface.co/MBZUAI/swiftformer-xs) on an unknown dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.6423
|
20 |
+
- Accuracy: 0.8165
|
21 |
|
22 |
## Model description
|
23 |
|
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
"epoch": 36.13,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime": 0.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
"total_flos": 9.686412043576934e+16,
|
9 |
-
"train_loss": 0.
|
10 |
-
"train_runtime":
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second": 1.
|
13 |
}
|
|
|
1 |
{
|
2 |
"epoch": 36.13,
|
3 |
+
"eval_accuracy": 0.8165137614678899,
|
4 |
+
"eval_loss": 0.6422722935676575,
|
5 |
+
"eval_runtime": 0.6417,
|
6 |
+
"eval_samples_per_second": 169.866,
|
7 |
+
"eval_steps_per_second": 6.234,
|
8 |
"total_flos": 9.686412043576934e+16,
|
9 |
+
"train_loss": 0.8966831156185695,
|
10 |
+
"train_runtime": 256.9851,
|
11 |
+
"train_samples_per_second": 152.227,
|
12 |
+
"train_steps_per_second": 1.09
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 36.13,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime": 0.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 36.13,
|
3 |
+
"eval_accuracy": 0.8165137614678899,
|
4 |
+
"eval_loss": 0.6422722935676575,
|
5 |
+
"eval_runtime": 0.6417,
|
6 |
+
"eval_samples_per_second": 169.866,
|
7 |
+
"eval_steps_per_second": 6.234
|
8 |
}
|
runs/Feb01_19-16-14_b2db6ba6b423/events.out.tfevents.1706815248.b2db6ba6b423.3854.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1e3e5b9f999bf8a5daf24051f68adfbe47da2a1f3c2cf2ba8c778853b2e31bc9
|
3 |
+
size 411
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 36.13,
|
3 |
"total_flos": 9.686412043576934e+16,
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 1.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 36.13,
|
3 |
"total_flos": 9.686412043576934e+16,
|
4 |
+
"train_loss": 0.8966831156185695,
|
5 |
+
"train_runtime": 256.9851,
|
6 |
+
"train_samples_per_second": 152.227,
|
7 |
+
"train_steps_per_second": 1.09
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "swiftformer-xs-dmae-va-U-40/checkpoint-
|
4 |
"epoch": 36.12903225806452,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 280,
|
@@ -10,513 +10,513 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.9,
|
13 |
-
"eval_accuracy": 0.
|
14 |
-
"eval_loss": 1.
|
15 |
-
"eval_runtime": 0.
|
16 |
-
"eval_samples_per_second":
|
17 |
-
"eval_steps_per_second":
|
18 |
"step": 7
|
19 |
},
|
20 |
{
|
21 |
"epoch": 1.29,
|
22 |
"learning_rate": 1.785714285714286e-05,
|
23 |
-
"loss": 1.
|
24 |
"step": 10
|
25 |
},
|
26 |
{
|
27 |
"epoch": 1.94,
|
28 |
-
"eval_accuracy": 0.
|
29 |
-
"eval_loss": 1.
|
30 |
-
"eval_runtime": 0.
|
31 |
-
"eval_samples_per_second":
|
32 |
-
"eval_steps_per_second":
|
33 |
"step": 15
|
34 |
},
|
35 |
{
|
36 |
"epoch": 2.58,
|
37 |
"learning_rate": 3.571428571428572e-05,
|
38 |
-
"loss": 1.
|
39 |
"step": 20
|
40 |
},
|
41 |
{
|
42 |
"epoch": 2.97,
|
43 |
-
"eval_accuracy": 0.
|
44 |
-
"eval_loss": 1.
|
45 |
-
"eval_runtime": 0.
|
46 |
-
"eval_samples_per_second":
|
47 |
-
"eval_steps_per_second":
|
48 |
"step": 23
|
49 |
},
|
50 |
{
|
51 |
"epoch": 3.87,
|
52 |
"learning_rate": 4.960317460317461e-05,
|
53 |
-
"loss": 1.
|
54 |
"step": 30
|
55 |
},
|
56 |
{
|
57 |
"epoch": 4.0,
|
58 |
-
"eval_accuracy": 0.
|
59 |
-
"eval_loss": 1.
|
60 |
-
"eval_runtime": 0.
|
61 |
-
"eval_samples_per_second":
|
62 |
-
"eval_steps_per_second":
|
63 |
"step": 31
|
64 |
},
|
65 |
{
|
66 |
"epoch": 4.9,
|
67 |
-
"eval_accuracy": 0.
|
68 |
-
"eval_loss": 1.
|
69 |
-
"eval_runtime": 0.
|
70 |
-
"eval_samples_per_second":
|
71 |
-
"eval_steps_per_second": 9.
|
72 |
"step": 38
|
73 |
},
|
74 |
{
|
75 |
"epoch": 5.16,
|
76 |
"learning_rate": 4.761904761904762e-05,
|
77 |
-
"loss": 1.
|
78 |
"step": 40
|
79 |
},
|
80 |
{
|
81 |
"epoch": 5.94,
|
82 |
-
"eval_accuracy": 0.
|
83 |
-
"eval_loss": 1.
|
84 |
-
"eval_runtime": 0.
|
85 |
-
"eval_samples_per_second":
|
86 |
-
"eval_steps_per_second":
|
87 |
"step": 46
|
88 |
},
|
89 |
{
|
90 |
"epoch": 6.45,
|
91 |
"learning_rate": 4.563492063492064e-05,
|
92 |
-
"loss": 1.
|
93 |
"step": 50
|
94 |
},
|
95 |
{
|
96 |
"epoch": 6.97,
|
97 |
-
"eval_accuracy": 0.
|
98 |
-
"eval_loss": 1.
|
99 |
-
"eval_runtime": 0.
|
100 |
-
"eval_samples_per_second":
|
101 |
-
"eval_steps_per_second":
|
102 |
"step": 54
|
103 |
},
|
104 |
{
|
105 |
"epoch": 7.74,
|
106 |
"learning_rate": 4.3650793650793655e-05,
|
107 |
-
"loss": 1.
|
108 |
"step": 60
|
109 |
},
|
110 |
{
|
111 |
"epoch": 8.0,
|
112 |
-
"eval_accuracy": 0.
|
113 |
-
"eval_loss": 1.
|
114 |
-
"eval_runtime": 0.
|
115 |
-
"eval_samples_per_second":
|
116 |
-
"eval_steps_per_second": 9.
|
117 |
"step": 62
|
118 |
},
|
119 |
{
|
120 |
"epoch": 8.9,
|
121 |
-
"eval_accuracy": 0.
|
122 |
-
"eval_loss":
|
123 |
-
"eval_runtime": 0.
|
124 |
-
"eval_samples_per_second":
|
125 |
-
"eval_steps_per_second": 7.
|
126 |
"step": 69
|
127 |
},
|
128 |
{
|
129 |
"epoch": 9.03,
|
130 |
"learning_rate": 4.166666666666667e-05,
|
131 |
-
"loss": 1.
|
132 |
"step": 70
|
133 |
},
|
134 |
{
|
135 |
"epoch": 9.94,
|
136 |
-
"eval_accuracy": 0.
|
137 |
-
"eval_loss": 0.
|
138 |
-
"eval_runtime": 0.
|
139 |
-
"eval_samples_per_second":
|
140 |
-
"eval_steps_per_second":
|
141 |
"step": 77
|
142 |
},
|
143 |
{
|
144 |
"epoch": 10.32,
|
145 |
"learning_rate": 3.968253968253968e-05,
|
146 |
-
"loss":
|
147 |
"step": 80
|
148 |
},
|
149 |
{
|
150 |
"epoch": 10.97,
|
151 |
-
"eval_accuracy": 0.
|
152 |
-
"eval_loss": 0.
|
153 |
-
"eval_runtime": 0.
|
154 |
-
"eval_samples_per_second":
|
155 |
-
"eval_steps_per_second":
|
156 |
"step": 85
|
157 |
},
|
158 |
{
|
159 |
"epoch": 11.61,
|
160 |
"learning_rate": 3.76984126984127e-05,
|
161 |
-
"loss": 0.
|
162 |
"step": 90
|
163 |
},
|
164 |
{
|
165 |
"epoch": 12.0,
|
166 |
-
"eval_accuracy": 0.
|
167 |
-
"eval_loss": 0.
|
168 |
-
"eval_runtime": 0.
|
169 |
-
"eval_samples_per_second":
|
170 |
-
"eval_steps_per_second":
|
171 |
"step": 93
|
172 |
},
|
173 |
{
|
174 |
"epoch": 12.9,
|
175 |
"learning_rate": 3.571428571428572e-05,
|
176 |
-
"loss": 0.
|
177 |
"step": 100
|
178 |
},
|
179 |
{
|
180 |
"epoch": 12.9,
|
181 |
-
"eval_accuracy": 0.
|
182 |
-
"eval_loss": 0.
|
183 |
-
"eval_runtime": 0.
|
184 |
-
"eval_samples_per_second":
|
185 |
-
"eval_steps_per_second":
|
186 |
"step": 100
|
187 |
},
|
188 |
{
|
189 |
"epoch": 13.94,
|
190 |
-
"eval_accuracy": 0.
|
191 |
-
"eval_loss": 0.
|
192 |
-
"eval_runtime": 0.
|
193 |
-
"eval_samples_per_second":
|
194 |
-
"eval_steps_per_second":
|
195 |
"step": 108
|
196 |
},
|
197 |
{
|
198 |
"epoch": 14.19,
|
199 |
"learning_rate": 3.3730158730158734e-05,
|
200 |
-
"loss": 0.
|
201 |
"step": 110
|
202 |
},
|
203 |
{
|
204 |
"epoch": 14.97,
|
205 |
"eval_accuracy": 0.7155963302752294,
|
206 |
-
"eval_loss": 0.
|
207 |
-
"eval_runtime": 0.
|
208 |
-
"eval_samples_per_second":
|
209 |
-
"eval_steps_per_second": 9.
|
210 |
"step": 116
|
211 |
},
|
212 |
{
|
213 |
"epoch": 15.48,
|
214 |
"learning_rate": 3.1746031746031745e-05,
|
215 |
-
"loss": 0.
|
216 |
"step": 120
|
217 |
},
|
218 |
{
|
219 |
"epoch": 16.0,
|
220 |
-
"eval_accuracy": 0.
|
221 |
-
"eval_loss": 0.
|
222 |
-
"eval_runtime": 0.
|
223 |
-
"eval_samples_per_second":
|
224 |
-
"eval_steps_per_second":
|
225 |
"step": 124
|
226 |
},
|
227 |
{
|
228 |
"epoch": 16.77,
|
229 |
"learning_rate": 2.9761904761904762e-05,
|
230 |
-
"loss": 0.
|
231 |
"step": 130
|
232 |
},
|
233 |
{
|
234 |
"epoch": 16.9,
|
235 |
"eval_accuracy": 0.6972477064220184,
|
236 |
-
"eval_loss": 0.
|
237 |
-
"eval_runtime": 0.
|
238 |
-
"eval_samples_per_second":
|
239 |
-
"eval_steps_per_second": 9.
|
240 |
"step": 131
|
241 |
},
|
242 |
{
|
243 |
"epoch": 17.94,
|
244 |
"eval_accuracy": 0.7339449541284404,
|
245 |
-
"eval_loss": 0.
|
246 |
-
"eval_runtime": 0.
|
247 |
-
"eval_samples_per_second":
|
248 |
-
"eval_steps_per_second":
|
249 |
"step": 139
|
250 |
},
|
251 |
{
|
252 |
"epoch": 18.06,
|
253 |
"learning_rate": 2.777777777777778e-05,
|
254 |
-
"loss": 0.
|
255 |
"step": 140
|
256 |
},
|
257 |
{
|
258 |
"epoch": 18.97,
|
259 |
-
"eval_accuracy": 0.
|
260 |
-
"eval_loss": 0.
|
261 |
-
"eval_runtime": 0.
|
262 |
-
"eval_samples_per_second":
|
263 |
-
"eval_steps_per_second": 9.
|
264 |
"step": 147
|
265 |
},
|
266 |
{
|
267 |
"epoch": 19.35,
|
268 |
"learning_rate": 2.5793650793650796e-05,
|
269 |
-
"loss": 0.
|
270 |
"step": 150
|
271 |
},
|
272 |
{
|
273 |
"epoch": 20.0,
|
274 |
-
"eval_accuracy": 0.
|
275 |
-
"eval_loss": 0.
|
276 |
-
"eval_runtime": 0.
|
277 |
-
"eval_samples_per_second":
|
278 |
-
"eval_steps_per_second": 9.
|
279 |
"step": 155
|
280 |
},
|
281 |
{
|
282 |
"epoch": 20.65,
|
283 |
"learning_rate": 2.380952380952381e-05,
|
284 |
-
"loss": 0.
|
285 |
"step": 160
|
286 |
},
|
287 |
{
|
288 |
"epoch": 20.9,
|
289 |
"eval_accuracy": 0.7431192660550459,
|
290 |
-
"eval_loss": 0.
|
291 |
-
"eval_runtime": 0.
|
292 |
-
"eval_samples_per_second":
|
293 |
-
"eval_steps_per_second":
|
294 |
"step": 162
|
295 |
},
|
296 |
{
|
297 |
"epoch": 21.94,
|
298 |
"learning_rate": 2.1825396825396827e-05,
|
299 |
-
"loss": 0.
|
300 |
"step": 170
|
301 |
},
|
302 |
{
|
303 |
"epoch": 21.94,
|
304 |
-
"eval_accuracy": 0.
|
305 |
-
"eval_loss": 0.
|
306 |
-
"eval_runtime": 0.
|
307 |
-
"eval_samples_per_second":
|
308 |
-
"eval_steps_per_second": 9.
|
309 |
"step": 170
|
310 |
},
|
311 |
{
|
312 |
"epoch": 22.97,
|
313 |
-
"eval_accuracy": 0.
|
314 |
-
"eval_loss": 0.
|
315 |
-
"eval_runtime": 0.
|
316 |
-
"eval_samples_per_second":
|
317 |
-
"eval_steps_per_second":
|
318 |
"step": 178
|
319 |
},
|
320 |
{
|
321 |
"epoch": 23.23,
|
322 |
"learning_rate": 1.984126984126984e-05,
|
323 |
-
"loss": 0.
|
324 |
"step": 180
|
325 |
},
|
326 |
{
|
327 |
"epoch": 24.0,
|
328 |
-
"eval_accuracy": 0.
|
329 |
-
"eval_loss": 0.
|
330 |
-
"eval_runtime": 0.
|
331 |
-
"eval_samples_per_second":
|
332 |
-
"eval_steps_per_second": 9.
|
333 |
"step": 186
|
334 |
},
|
335 |
{
|
336 |
"epoch": 24.52,
|
337 |
"learning_rate": 1.785714285714286e-05,
|
338 |
-
"loss": 0.
|
339 |
"step": 190
|
340 |
},
|
341 |
{
|
342 |
"epoch": 24.9,
|
343 |
-
"eval_accuracy": 0.
|
344 |
-
"eval_loss": 0.
|
345 |
-
"eval_runtime": 0.
|
346 |
-
"eval_samples_per_second":
|
347 |
-
"eval_steps_per_second": 9.
|
348 |
"step": 193
|
349 |
},
|
350 |
{
|
351 |
"epoch": 25.81,
|
352 |
"learning_rate": 1.5873015873015872e-05,
|
353 |
-
"loss": 0.
|
354 |
"step": 200
|
355 |
},
|
356 |
{
|
357 |
"epoch": 25.94,
|
358 |
-
"eval_accuracy": 0.
|
359 |
-
"eval_loss": 0.
|
360 |
-
"eval_runtime": 0.
|
361 |
-
"eval_samples_per_second":
|
362 |
-
"eval_steps_per_second": 9.
|
363 |
"step": 201
|
364 |
},
|
365 |
{
|
366 |
"epoch": 26.97,
|
367 |
-
"eval_accuracy": 0.
|
368 |
-
"eval_loss": 0.
|
369 |
-
"eval_runtime": 0.
|
370 |
-
"eval_samples_per_second":
|
371 |
-
"eval_steps_per_second":
|
372 |
"step": 209
|
373 |
},
|
374 |
{
|
375 |
"epoch": 27.1,
|
376 |
"learning_rate": 1.388888888888889e-05,
|
377 |
-
"loss": 0.
|
378 |
"step": 210
|
379 |
},
|
380 |
{
|
381 |
"epoch": 28.0,
|
382 |
-
"eval_accuracy": 0.
|
383 |
-
"eval_loss": 0.
|
384 |
-
"eval_runtime": 0.
|
385 |
-
"eval_samples_per_second":
|
386 |
-
"eval_steps_per_second":
|
387 |
"step": 217
|
388 |
},
|
389 |
{
|
390 |
"epoch": 28.39,
|
391 |
"learning_rate": 1.1904761904761905e-05,
|
392 |
-
"loss": 0.
|
393 |
"step": 220
|
394 |
},
|
395 |
{
|
396 |
"epoch": 28.9,
|
397 |
-
"eval_accuracy": 0.
|
398 |
-
"eval_loss": 0.
|
399 |
-
"eval_runtime": 0.
|
400 |
-
"eval_samples_per_second":
|
401 |
-
"eval_steps_per_second": 9.
|
402 |
"step": 224
|
403 |
},
|
404 |
{
|
405 |
"epoch": 29.68,
|
406 |
"learning_rate": 9.92063492063492e-06,
|
407 |
-
"loss": 0.
|
408 |
"step": 230
|
409 |
},
|
410 |
{
|
411 |
"epoch": 29.94,
|
412 |
-
"eval_accuracy": 0.
|
413 |
-
"eval_loss": 0.
|
414 |
-
"eval_runtime": 0.
|
415 |
-
"eval_samples_per_second":
|
416 |
-
"eval_steps_per_second":
|
417 |
"step": 232
|
418 |
},
|
419 |
{
|
420 |
"epoch": 30.97,
|
421 |
"learning_rate": 7.936507936507936e-06,
|
422 |
-
"loss": 0.
|
423 |
"step": 240
|
424 |
},
|
425 |
{
|
426 |
"epoch": 30.97,
|
427 |
-
"eval_accuracy": 0.
|
428 |
-
"eval_loss": 0.
|
429 |
-
"eval_runtime": 0.
|
430 |
-
"eval_samples_per_second":
|
431 |
-
"eval_steps_per_second": 9.
|
432 |
"step": 240
|
433 |
},
|
434 |
{
|
435 |
"epoch": 32.0,
|
436 |
-
"eval_accuracy": 0.
|
437 |
-
"eval_loss": 0.
|
438 |
-
"eval_runtime": 0.
|
439 |
-
"eval_samples_per_second":
|
440 |
-
"eval_steps_per_second":
|
441 |
"step": 248
|
442 |
},
|
443 |
{
|
444 |
"epoch": 32.26,
|
445 |
"learning_rate": 5.9523809523809525e-06,
|
446 |
-
"loss": 0.
|
447 |
"step": 250
|
448 |
},
|
449 |
{
|
450 |
"epoch": 32.9,
|
451 |
-
"eval_accuracy": 0.
|
452 |
-
"eval_loss": 0.
|
453 |
-
"eval_runtime": 0.
|
454 |
-
"eval_samples_per_second":
|
455 |
-
"eval_steps_per_second": 9.
|
456 |
"step": 255
|
457 |
},
|
458 |
{
|
459 |
"epoch": 33.55,
|
460 |
"learning_rate": 3.968253968253968e-06,
|
461 |
-
"loss": 0.
|
462 |
"step": 260
|
463 |
},
|
464 |
{
|
465 |
"epoch": 33.94,
|
466 |
-
"eval_accuracy": 0.
|
467 |
-
"eval_loss": 0.
|
468 |
-
"eval_runtime": 0.
|
469 |
-
"eval_samples_per_second": 259.
|
470 |
-
"eval_steps_per_second": 9.
|
471 |
"step": 263
|
472 |
},
|
473 |
{
|
474 |
"epoch": 34.84,
|
475 |
"learning_rate": 1.984126984126984e-06,
|
476 |
-
"loss": 0.
|
477 |
"step": 270
|
478 |
},
|
479 |
{
|
480 |
"epoch": 34.97,
|
481 |
-
"eval_accuracy": 0.
|
482 |
-
"eval_loss": 0.
|
483 |
-
"eval_runtime": 0.
|
484 |
-
"eval_samples_per_second":
|
485 |
-
"eval_steps_per_second":
|
486 |
"step": 271
|
487 |
},
|
488 |
{
|
489 |
"epoch": 36.0,
|
490 |
-
"eval_accuracy": 0.
|
491 |
-
"eval_loss": 0.
|
492 |
-
"eval_runtime": 0.
|
493 |
-
"eval_samples_per_second":
|
494 |
-
"eval_steps_per_second":
|
495 |
"step": 279
|
496 |
},
|
497 |
{
|
498 |
"epoch": 36.13,
|
499 |
"learning_rate": 0.0,
|
500 |
-
"loss": 0.
|
501 |
"step": 280
|
502 |
},
|
503 |
{
|
504 |
"epoch": 36.13,
|
505 |
-
"eval_accuracy": 0.
|
506 |
-
"eval_loss": 0.
|
507 |
-
"eval_runtime": 0.
|
508 |
-
"eval_samples_per_second":
|
509 |
-
"eval_steps_per_second":
|
510 |
"step": 280
|
511 |
},
|
512 |
{
|
513 |
"epoch": 36.13,
|
514 |
"step": 280,
|
515 |
"total_flos": 9.686412043576934e+16,
|
516 |
-
"train_loss": 0.
|
517 |
-
"train_runtime":
|
518 |
-
"train_samples_per_second":
|
519 |
-
"train_steps_per_second": 1.
|
520 |
}
|
521 |
],
|
522 |
"logging_steps": 10,
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.8165137614678899,
|
3 |
+
"best_model_checkpoint": "swiftformer-xs-dmae-va-U-40/checkpoint-232",
|
4 |
"epoch": 36.12903225806452,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 280,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.9,
|
13 |
+
"eval_accuracy": 0.3211009174311927,
|
14 |
+
"eval_loss": 1.3882640600204468,
|
15 |
+
"eval_runtime": 0.5018,
|
16 |
+
"eval_samples_per_second": 217.232,
|
17 |
+
"eval_steps_per_second": 7.972,
|
18 |
"step": 7
|
19 |
},
|
20 |
{
|
21 |
"epoch": 1.29,
|
22 |
"learning_rate": 1.785714285714286e-05,
|
23 |
+
"loss": 1.4011,
|
24 |
"step": 10
|
25 |
},
|
26 |
{
|
27 |
"epoch": 1.94,
|
28 |
+
"eval_accuracy": 0.3577981651376147,
|
29 |
+
"eval_loss": 1.3383492231369019,
|
30 |
+
"eval_runtime": 0.5683,
|
31 |
+
"eval_samples_per_second": 191.814,
|
32 |
+
"eval_steps_per_second": 7.039,
|
33 |
"step": 15
|
34 |
},
|
35 |
{
|
36 |
"epoch": 2.58,
|
37 |
"learning_rate": 3.571428571428572e-05,
|
38 |
+
"loss": 1.3646,
|
39 |
"step": 20
|
40 |
},
|
41 |
{
|
42 |
"epoch": 2.97,
|
43 |
+
"eval_accuracy": 0.44036697247706424,
|
44 |
+
"eval_loss": 1.280238151550293,
|
45 |
+
"eval_runtime": 0.4561,
|
46 |
+
"eval_samples_per_second": 238.958,
|
47 |
+
"eval_steps_per_second": 8.769,
|
48 |
"step": 23
|
49 |
},
|
50 |
{
|
51 |
"epoch": 3.87,
|
52 |
"learning_rate": 4.960317460317461e-05,
|
53 |
+
"loss": 1.315,
|
54 |
"step": 30
|
55 |
},
|
56 |
{
|
57 |
"epoch": 4.0,
|
58 |
+
"eval_accuracy": 0.44954128440366975,
|
59 |
+
"eval_loss": 1.2193504571914673,
|
60 |
+
"eval_runtime": 0.8508,
|
61 |
+
"eval_samples_per_second": 128.117,
|
62 |
+
"eval_steps_per_second": 4.702,
|
63 |
"step": 31
|
64 |
},
|
65 |
{
|
66 |
"epoch": 4.9,
|
67 |
+
"eval_accuracy": 0.5229357798165137,
|
68 |
+
"eval_loss": 1.1717596054077148,
|
69 |
+
"eval_runtime": 0.4255,
|
70 |
+
"eval_samples_per_second": 256.17,
|
71 |
+
"eval_steps_per_second": 9.401,
|
72 |
"step": 38
|
73 |
},
|
74 |
{
|
75 |
"epoch": 5.16,
|
76 |
"learning_rate": 4.761904761904762e-05,
|
77 |
+
"loss": 1.2634,
|
78 |
"step": 40
|
79 |
},
|
80 |
{
|
81 |
"epoch": 5.94,
|
82 |
+
"eval_accuracy": 0.5045871559633027,
|
83 |
+
"eval_loss": 1.127877950668335,
|
84 |
+
"eval_runtime": 0.4475,
|
85 |
+
"eval_samples_per_second": 243.601,
|
86 |
+
"eval_steps_per_second": 8.939,
|
87 |
"step": 46
|
88 |
},
|
89 |
{
|
90 |
"epoch": 6.45,
|
91 |
"learning_rate": 4.563492063492064e-05,
|
92 |
+
"loss": 1.1949,
|
93 |
"step": 50
|
94 |
},
|
95 |
{
|
96 |
"epoch": 6.97,
|
97 |
+
"eval_accuracy": 0.5871559633027523,
|
98 |
+
"eval_loss": 1.0761113166809082,
|
99 |
+
"eval_runtime": 0.5058,
|
100 |
+
"eval_samples_per_second": 215.497,
|
101 |
+
"eval_steps_per_second": 7.908,
|
102 |
"step": 54
|
103 |
},
|
104 |
{
|
105 |
"epoch": 7.74,
|
106 |
"learning_rate": 4.3650793650793655e-05,
|
107 |
+
"loss": 1.1136,
|
108 |
"step": 60
|
109 |
},
|
110 |
{
|
111 |
"epoch": 8.0,
|
112 |
+
"eval_accuracy": 0.6330275229357798,
|
113 |
+
"eval_loss": 1.0223767757415771,
|
114 |
+
"eval_runtime": 0.4226,
|
115 |
+
"eval_samples_per_second": 257.914,
|
116 |
+
"eval_steps_per_second": 9.465,
|
117 |
"step": 62
|
118 |
},
|
119 |
{
|
120 |
"epoch": 8.9,
|
121 |
+
"eval_accuracy": 0.6238532110091743,
|
122 |
+
"eval_loss": 0.9975973963737488,
|
123 |
+
"eval_runtime": 0.556,
|
124 |
+
"eval_samples_per_second": 196.055,
|
125 |
+
"eval_steps_per_second": 7.195,
|
126 |
"step": 69
|
127 |
},
|
128 |
{
|
129 |
"epoch": 9.03,
|
130 |
"learning_rate": 4.166666666666667e-05,
|
131 |
+
"loss": 1.0824,
|
132 |
"step": 70
|
133 |
},
|
134 |
{
|
135 |
"epoch": 9.94,
|
136 |
+
"eval_accuracy": 0.6605504587155964,
|
137 |
+
"eval_loss": 0.9517724514007568,
|
138 |
+
"eval_runtime": 0.4569,
|
139 |
+
"eval_samples_per_second": 238.566,
|
140 |
+
"eval_steps_per_second": 8.755,
|
141 |
"step": 77
|
142 |
},
|
143 |
{
|
144 |
"epoch": 10.32,
|
145 |
"learning_rate": 3.968253968253968e-05,
|
146 |
+
"loss": 1.0212,
|
147 |
"step": 80
|
148 |
},
|
149 |
{
|
150 |
"epoch": 10.97,
|
151 |
+
"eval_accuracy": 0.6697247706422018,
|
152 |
+
"eval_loss": 0.9116750359535217,
|
153 |
+
"eval_runtime": 0.4487,
|
154 |
+
"eval_samples_per_second": 242.931,
|
155 |
+
"eval_steps_per_second": 8.915,
|
156 |
"step": 85
|
157 |
},
|
158 |
{
|
159 |
"epoch": 11.61,
|
160 |
"learning_rate": 3.76984126984127e-05,
|
161 |
+
"loss": 0.9566,
|
162 |
"step": 90
|
163 |
},
|
164 |
{
|
165 |
"epoch": 12.0,
|
166 |
+
"eval_accuracy": 0.6880733944954128,
|
167 |
+
"eval_loss": 0.8973050117492676,
|
168 |
+
"eval_runtime": 0.448,
|
169 |
+
"eval_samples_per_second": 243.303,
|
170 |
+
"eval_steps_per_second": 8.929,
|
171 |
"step": 93
|
172 |
},
|
173 |
{
|
174 |
"epoch": 12.9,
|
175 |
"learning_rate": 3.571428571428572e-05,
|
176 |
+
"loss": 0.935,
|
177 |
"step": 100
|
178 |
},
|
179 |
{
|
180 |
"epoch": 12.9,
|
181 |
+
"eval_accuracy": 0.7064220183486238,
|
182 |
+
"eval_loss": 0.8704947233200073,
|
183 |
+
"eval_runtime": 0.4696,
|
184 |
+
"eval_samples_per_second": 232.132,
|
185 |
+
"eval_steps_per_second": 8.519,
|
186 |
"step": 100
|
187 |
},
|
188 |
{
|
189 |
"epoch": 13.94,
|
190 |
+
"eval_accuracy": 0.7155963302752294,
|
191 |
+
"eval_loss": 0.8559499382972717,
|
192 |
+
"eval_runtime": 0.5932,
|
193 |
+
"eval_samples_per_second": 183.74,
|
194 |
+
"eval_steps_per_second": 6.743,
|
195 |
"step": 108
|
196 |
},
|
197 |
{
|
198 |
"epoch": 14.19,
|
199 |
"learning_rate": 3.3730158730158734e-05,
|
200 |
+
"loss": 0.8826,
|
201 |
"step": 110
|
202 |
},
|
203 |
{
|
204 |
"epoch": 14.97,
|
205 |
"eval_accuracy": 0.7155963302752294,
|
206 |
+
"eval_loss": 0.8371049165725708,
|
207 |
+
"eval_runtime": 0.4298,
|
208 |
+
"eval_samples_per_second": 253.624,
|
209 |
+
"eval_steps_per_second": 9.307,
|
210 |
"step": 116
|
211 |
},
|
212 |
{
|
213 |
"epoch": 15.48,
|
214 |
"learning_rate": 3.1746031746031745e-05,
|
215 |
+
"loss": 0.8688,
|
216 |
"step": 120
|
217 |
},
|
218 |
{
|
219 |
"epoch": 16.0,
|
220 |
+
"eval_accuracy": 0.7155963302752294,
|
221 |
+
"eval_loss": 0.8251588940620422,
|
222 |
+
"eval_runtime": 0.4656,
|
223 |
+
"eval_samples_per_second": 234.09,
|
224 |
+
"eval_steps_per_second": 8.59,
|
225 |
"step": 124
|
226 |
},
|
227 |
{
|
228 |
"epoch": 16.77,
|
229 |
"learning_rate": 2.9761904761904762e-05,
|
230 |
+
"loss": 0.8436,
|
231 |
"step": 130
|
232 |
},
|
233 |
{
|
234 |
"epoch": 16.9,
|
235 |
"eval_accuracy": 0.6972477064220184,
|
236 |
+
"eval_loss": 0.8211333155632019,
|
237 |
+
"eval_runtime": 0.4255,
|
238 |
+
"eval_samples_per_second": 256.176,
|
239 |
+
"eval_steps_per_second": 9.401,
|
240 |
"step": 131
|
241 |
},
|
242 |
{
|
243 |
"epoch": 17.94,
|
244 |
"eval_accuracy": 0.7339449541284404,
|
245 |
+
"eval_loss": 0.8039615750312805,
|
246 |
+
"eval_runtime": 0.4375,
|
247 |
+
"eval_samples_per_second": 249.151,
|
248 |
+
"eval_steps_per_second": 9.143,
|
249 |
"step": 139
|
250 |
},
|
251 |
{
|
252 |
"epoch": 18.06,
|
253 |
"learning_rate": 2.777777777777778e-05,
|
254 |
+
"loss": 0.8155,
|
255 |
"step": 140
|
256 |
},
|
257 |
{
|
258 |
"epoch": 18.97,
|
259 |
+
"eval_accuracy": 0.7431192660550459,
|
260 |
+
"eval_loss": 0.7625071406364441,
|
261 |
+
"eval_runtime": 0.435,
|
262 |
+
"eval_samples_per_second": 250.55,
|
263 |
+
"eval_steps_per_second": 9.195,
|
264 |
"step": 147
|
265 |
},
|
266 |
{
|
267 |
"epoch": 19.35,
|
268 |
"learning_rate": 2.5793650793650796e-05,
|
269 |
+
"loss": 0.7831,
|
270 |
"step": 150
|
271 |
},
|
272 |
{
|
273 |
"epoch": 20.0,
|
274 |
+
"eval_accuracy": 0.7431192660550459,
|
275 |
+
"eval_loss": 0.7451765537261963,
|
276 |
+
"eval_runtime": 0.4363,
|
277 |
+
"eval_samples_per_second": 249.828,
|
278 |
+
"eval_steps_per_second": 9.168,
|
279 |
"step": 155
|
280 |
},
|
281 |
{
|
282 |
"epoch": 20.65,
|
283 |
"learning_rate": 2.380952380952381e-05,
|
284 |
+
"loss": 0.7826,
|
285 |
"step": 160
|
286 |
},
|
287 |
{
|
288 |
"epoch": 20.9,
|
289 |
"eval_accuracy": 0.7431192660550459,
|
290 |
+
"eval_loss": 0.7278565764427185,
|
291 |
+
"eval_runtime": 0.5078,
|
292 |
+
"eval_samples_per_second": 214.643,
|
293 |
+
"eval_steps_per_second": 7.877,
|
294 |
"step": 162
|
295 |
},
|
296 |
{
|
297 |
"epoch": 21.94,
|
298 |
"learning_rate": 2.1825396825396827e-05,
|
299 |
+
"loss": 0.7499,
|
300 |
"step": 170
|
301 |
},
|
302 |
{
|
303 |
"epoch": 21.94,
|
304 |
+
"eval_accuracy": 0.7431192660550459,
|
305 |
+
"eval_loss": 0.714833676815033,
|
306 |
+
"eval_runtime": 0.4399,
|
307 |
+
"eval_samples_per_second": 247.796,
|
308 |
+
"eval_steps_per_second": 9.093,
|
309 |
"step": 170
|
310 |
},
|
311 |
{
|
312 |
"epoch": 22.97,
|
313 |
+
"eval_accuracy": 0.7522935779816514,
|
314 |
+
"eval_loss": 0.7061465382575989,
|
315 |
+
"eval_runtime": 0.5513,
|
316 |
+
"eval_samples_per_second": 197.698,
|
317 |
+
"eval_steps_per_second": 7.255,
|
318 |
"step": 178
|
319 |
},
|
320 |
{
|
321 |
"epoch": 23.23,
|
322 |
"learning_rate": 1.984126984126984e-05,
|
323 |
+
"loss": 0.7539,
|
324 |
"step": 180
|
325 |
},
|
326 |
{
|
327 |
"epoch": 24.0,
|
328 |
+
"eval_accuracy": 0.7522935779816514,
|
329 |
+
"eval_loss": 0.7026045918464661,
|
330 |
+
"eval_runtime": 0.4348,
|
331 |
+
"eval_samples_per_second": 250.669,
|
332 |
+
"eval_steps_per_second": 9.199,
|
333 |
"step": 186
|
334 |
},
|
335 |
{
|
336 |
"epoch": 24.52,
|
337 |
"learning_rate": 1.785714285714286e-05,
|
338 |
+
"loss": 0.7453,
|
339 |
"step": 190
|
340 |
},
|
341 |
{
|
342 |
"epoch": 24.9,
|
343 |
+
"eval_accuracy": 0.7889908256880734,
|
344 |
+
"eval_loss": 0.6818734407424927,
|
345 |
+
"eval_runtime": 0.4308,
|
346 |
+
"eval_samples_per_second": 253.004,
|
347 |
+
"eval_steps_per_second": 9.285,
|
348 |
"step": 193
|
349 |
},
|
350 |
{
|
351 |
"epoch": 25.81,
|
352 |
"learning_rate": 1.5873015873015872e-05,
|
353 |
+
"loss": 0.7174,
|
354 |
"step": 200
|
355 |
},
|
356 |
{
|
357 |
"epoch": 25.94,
|
358 |
+
"eval_accuracy": 0.7706422018348624,
|
359 |
+
"eval_loss": 0.6837214231491089,
|
360 |
+
"eval_runtime": 0.4357,
|
361 |
+
"eval_samples_per_second": 250.193,
|
362 |
+
"eval_steps_per_second": 9.181,
|
363 |
"step": 201
|
364 |
},
|
365 |
{
|
366 |
"epoch": 26.97,
|
367 |
+
"eval_accuracy": 0.7798165137614679,
|
368 |
+
"eval_loss": 0.6743292212486267,
|
369 |
+
"eval_runtime": 0.4526,
|
370 |
+
"eval_samples_per_second": 240.854,
|
371 |
+
"eval_steps_per_second": 8.839,
|
372 |
"step": 209
|
373 |
},
|
374 |
{
|
375 |
"epoch": 27.1,
|
376 |
"learning_rate": 1.388888888888889e-05,
|
377 |
+
"loss": 0.7083,
|
378 |
"step": 210
|
379 |
},
|
380 |
{
|
381 |
"epoch": 28.0,
|
382 |
+
"eval_accuracy": 0.7798165137614679,
|
383 |
+
"eval_loss": 0.6706274151802063,
|
384 |
+
"eval_runtime": 0.4549,
|
385 |
+
"eval_samples_per_second": 239.637,
|
386 |
+
"eval_steps_per_second": 8.794,
|
387 |
"step": 217
|
388 |
},
|
389 |
{
|
390 |
"epoch": 28.39,
|
391 |
"learning_rate": 1.1904761904761905e-05,
|
392 |
+
"loss": 0.6813,
|
393 |
"step": 220
|
394 |
},
|
395 |
{
|
396 |
"epoch": 28.9,
|
397 |
+
"eval_accuracy": 0.8073394495412844,
|
398 |
+
"eval_loss": 0.6643755435943604,
|
399 |
+
"eval_runtime": 0.4318,
|
400 |
+
"eval_samples_per_second": 252.45,
|
401 |
+
"eval_steps_per_second": 9.264,
|
402 |
"step": 224
|
403 |
},
|
404 |
{
|
405 |
"epoch": 29.68,
|
406 |
"learning_rate": 9.92063492063492e-06,
|
407 |
+
"loss": 0.7107,
|
408 |
"step": 230
|
409 |
},
|
410 |
{
|
411 |
"epoch": 29.94,
|
412 |
+
"eval_accuracy": 0.8165137614678899,
|
413 |
+
"eval_loss": 0.6422722935676575,
|
414 |
+
"eval_runtime": 0.5633,
|
415 |
+
"eval_samples_per_second": 193.507,
|
416 |
+
"eval_steps_per_second": 7.101,
|
417 |
"step": 232
|
418 |
},
|
419 |
{
|
420 |
"epoch": 30.97,
|
421 |
"learning_rate": 7.936507936507936e-06,
|
422 |
+
"loss": 0.6912,
|
423 |
"step": 240
|
424 |
},
|
425 |
{
|
426 |
"epoch": 30.97,
|
427 |
+
"eval_accuracy": 0.7889908256880734,
|
428 |
+
"eval_loss": 0.6418679356575012,
|
429 |
+
"eval_runtime": 0.4378,
|
430 |
+
"eval_samples_per_second": 248.984,
|
431 |
+
"eval_steps_per_second": 9.137,
|
432 |
"step": 240
|
433 |
},
|
434 |
{
|
435 |
"epoch": 32.0,
|
436 |
+
"eval_accuracy": 0.7889908256880734,
|
437 |
+
"eval_loss": 0.6465409994125366,
|
438 |
+
"eval_runtime": 0.4513,
|
439 |
+
"eval_samples_per_second": 241.531,
|
440 |
+
"eval_steps_per_second": 8.864,
|
441 |
"step": 248
|
442 |
},
|
443 |
{
|
444 |
"epoch": 32.26,
|
445 |
"learning_rate": 5.9523809523809525e-06,
|
446 |
+
"loss": 0.7031,
|
447 |
"step": 250
|
448 |
},
|
449 |
{
|
450 |
"epoch": 32.9,
|
451 |
+
"eval_accuracy": 0.8073394495412844,
|
452 |
+
"eval_loss": 0.6346263289451599,
|
453 |
+
"eval_runtime": 0.4298,
|
454 |
+
"eval_samples_per_second": 253.596,
|
455 |
+
"eval_steps_per_second": 9.306,
|
456 |
"step": 255
|
457 |
},
|
458 |
{
|
459 |
"epoch": 33.55,
|
460 |
"learning_rate": 3.968253968253968e-06,
|
461 |
+
"loss": 0.6647,
|
462 |
"step": 260
|
463 |
},
|
464 |
{
|
465 |
"epoch": 33.94,
|
466 |
+
"eval_accuracy": 0.8073394495412844,
|
467 |
+
"eval_loss": 0.6346942186355591,
|
468 |
+
"eval_runtime": 0.4204,
|
469 |
+
"eval_samples_per_second": 259.297,
|
470 |
+
"eval_steps_per_second": 9.515,
|
471 |
"step": 263
|
472 |
},
|
473 |
{
|
474 |
"epoch": 34.84,
|
475 |
"learning_rate": 1.984126984126984e-06,
|
476 |
+
"loss": 0.6799,
|
477 |
"step": 270
|
478 |
},
|
479 |
{
|
480 |
"epoch": 34.97,
|
481 |
+
"eval_accuracy": 0.7981651376146789,
|
482 |
+
"eval_loss": 0.6475719213485718,
|
483 |
+
"eval_runtime": 0.5807,
|
484 |
+
"eval_samples_per_second": 187.701,
|
485 |
+
"eval_steps_per_second": 6.888,
|
486 |
"step": 271
|
487 |
},
|
488 |
{
|
489 |
"epoch": 36.0,
|
490 |
+
"eval_accuracy": 0.7981651376146789,
|
491 |
+
"eval_loss": 0.6428852081298828,
|
492 |
+
"eval_runtime": 0.4339,
|
493 |
+
"eval_samples_per_second": 251.222,
|
494 |
+
"eval_steps_per_second": 9.219,
|
495 |
"step": 279
|
496 |
},
|
497 |
{
|
498 |
"epoch": 36.13,
|
499 |
"learning_rate": 0.0,
|
500 |
+
"loss": 0.6774,
|
501 |
"step": 280
|
502 |
},
|
503 |
{
|
504 |
"epoch": 36.13,
|
505 |
+
"eval_accuracy": 0.7889908256880734,
|
506 |
+
"eval_loss": 0.6517751812934875,
|
507 |
+
"eval_runtime": 0.4506,
|
508 |
+
"eval_samples_per_second": 241.9,
|
509 |
+
"eval_steps_per_second": 8.877,
|
510 |
"step": 280
|
511 |
},
|
512 |
{
|
513 |
"epoch": 36.13,
|
514 |
"step": 280,
|
515 |
"total_flos": 9.686412043576934e+16,
|
516 |
+
"train_loss": 0.8966831156185695,
|
517 |
+
"train_runtime": 256.9851,
|
518 |
+
"train_samples_per_second": 152.227,
|
519 |
+
"train_steps_per_second": 1.09
|
520 |
}
|
521 |
],
|
522 |
"logging_steps": 10,
|