JEdward7777
commited on
Commit
•
fc756e8
1
Parent(s):
d2e4601
End of training
Browse files- all_results.json +4 -4
- train_results.json +4 -4
- trainer_state.json +216 -216
all_results.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"epoch": 50.0,
|
3 |
-
"train_loss":
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples_per_second":
|
6 |
-
"train_steps_per_second": 1.
|
7 |
}
|
|
|
1 |
{
|
2 |
"epoch": 50.0,
|
3 |
+
"train_loss": 23.064505024571574,
|
4 |
+
"train_runtime": 2425.9558,
|
5 |
+
"train_samples_per_second": 10.12,
|
6 |
+
"train_steps_per_second": 1.278
|
7 |
}
|
train_results.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"epoch": 50.0,
|
3 |
-
"train_loss":
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples_per_second":
|
6 |
-
"train_steps_per_second": 1.
|
7 |
}
|
|
|
1 |
{
|
2 |
"epoch": 50.0,
|
3 |
+
"train_loss": 23.064505024571574,
|
4 |
+
"train_runtime": 2425.9558,
|
5 |
+
"train_samples_per_second": 10.12,
|
6 |
+
"train_steps_per_second": 1.278
|
7 |
}
|
trainer_state.json
CHANGED
@@ -9,448 +9,448 @@
|
|
9 |
"log_history": [
|
10 |
{
|
11 |
"epoch": 1.0,
|
12 |
-
"eval_loss":
|
13 |
-
"eval_runtime":
|
14 |
-
"eval_samples_per_second":
|
15 |
-
"eval_steps_per_second": 5.
|
16 |
"step": 62
|
17 |
},
|
18 |
{
|
19 |
"epoch": 2.0,
|
20 |
-
"eval_loss": 39.
|
21 |
-
"eval_runtime":
|
22 |
-
"eval_samples_per_second":
|
23 |
-
"eval_steps_per_second":
|
24 |
"step": 124
|
25 |
},
|
26 |
{
|
27 |
"epoch": 3.0,
|
28 |
-
"eval_loss":
|
29 |
-
"eval_runtime": 3.
|
30 |
-
"eval_samples_per_second":
|
31 |
-
"eval_steps_per_second":
|
32 |
"step": 186
|
33 |
},
|
34 |
{
|
35 |
"epoch": 4.0,
|
36 |
-
"eval_loss":
|
37 |
-
"eval_runtime": 3.
|
38 |
-
"eval_samples_per_second":
|
39 |
-
"eval_steps_per_second":
|
40 |
"step": 248
|
41 |
},
|
42 |
{
|
43 |
"epoch": 5.0,
|
44 |
-
"eval_loss":
|
45 |
-
"eval_runtime": 3.
|
46 |
-
"eval_samples_per_second":
|
47 |
-
"eval_steps_per_second":
|
48 |
"step": 310
|
49 |
},
|
50 |
{
|
51 |
"epoch": 6.0,
|
52 |
-
"eval_loss":
|
53 |
-
"eval_runtime": 3.
|
54 |
-
"eval_samples_per_second":
|
55 |
-
"eval_steps_per_second":
|
56 |
"step": 372
|
57 |
},
|
58 |
{
|
59 |
"epoch": 7.0,
|
60 |
-
"eval_loss":
|
61 |
-
"eval_runtime": 3.
|
62 |
-
"eval_samples_per_second":
|
63 |
-
"eval_steps_per_second":
|
64 |
"step": 434
|
65 |
},
|
66 |
{
|
67 |
"epoch": 8.0,
|
68 |
-
"eval_loss":
|
69 |
-
"eval_runtime": 3.
|
70 |
-
"eval_samples_per_second":
|
71 |
-
"eval_steps_per_second":
|
72 |
"step": 496
|
73 |
},
|
74 |
{
|
75 |
"epoch": 8.06,
|
76 |
-
"learning_rate":
|
77 |
-
"loss":
|
78 |
"step": 500
|
79 |
},
|
80 |
{
|
81 |
"epoch": 9.0,
|
82 |
-
"eval_loss":
|
83 |
-
"eval_runtime": 3.
|
84 |
-
"eval_samples_per_second":
|
85 |
-
"eval_steps_per_second": 4.
|
86 |
"step": 558
|
87 |
},
|
88 |
{
|
89 |
"epoch": 10.0,
|
90 |
-
"eval_loss":
|
91 |
-
"eval_runtime": 3.
|
92 |
-
"eval_samples_per_second":
|
93 |
-
"eval_steps_per_second":
|
94 |
"step": 620
|
95 |
},
|
96 |
{
|
97 |
"epoch": 11.0,
|
98 |
-
"eval_loss":
|
99 |
-
"eval_runtime": 3.
|
100 |
-
"eval_samples_per_second":
|
101 |
-
"eval_steps_per_second":
|
102 |
"step": 682
|
103 |
},
|
104 |
{
|
105 |
"epoch": 12.0,
|
106 |
-
"eval_loss":
|
107 |
-
"eval_runtime": 3.
|
108 |
-
"eval_samples_per_second":
|
109 |
-
"eval_steps_per_second":
|
110 |
"step": 744
|
111 |
},
|
112 |
{
|
113 |
"epoch": 13.0,
|
114 |
-
"eval_loss":
|
115 |
-
"eval_runtime": 3.
|
116 |
-
"eval_samples_per_second":
|
117 |
-
"eval_steps_per_second":
|
118 |
"step": 806
|
119 |
},
|
120 |
{
|
121 |
"epoch": 14.0,
|
122 |
-
"eval_loss":
|
123 |
-
"eval_runtime": 3.
|
124 |
-
"eval_samples_per_second":
|
125 |
-
"eval_steps_per_second":
|
126 |
"step": 868
|
127 |
},
|
128 |
{
|
129 |
"epoch": 15.0,
|
130 |
-
"eval_loss":
|
131 |
-
"eval_runtime": 3.
|
132 |
-
"eval_samples_per_second":
|
133 |
-
"eval_steps_per_second":
|
134 |
"step": 930
|
135 |
},
|
136 |
{
|
137 |
"epoch": 16.0,
|
138 |
-
"eval_loss":
|
139 |
-
"eval_runtime": 3.
|
140 |
-
"eval_samples_per_second":
|
141 |
-
"eval_steps_per_second":
|
142 |
"step": 992
|
143 |
},
|
144 |
{
|
145 |
"epoch": 16.13,
|
146 |
-
"learning_rate":
|
147 |
-
"loss":
|
148 |
"step": 1000
|
149 |
},
|
150 |
{
|
151 |
"epoch": 17.0,
|
152 |
-
"eval_loss":
|
153 |
-
"eval_runtime": 3.
|
154 |
-
"eval_samples_per_second":
|
155 |
-
"eval_steps_per_second": 4.
|
156 |
"step": 1054
|
157 |
},
|
158 |
{
|
159 |
"epoch": 18.0,
|
160 |
-
"eval_loss":
|
161 |
-
"eval_runtime": 3.
|
162 |
-
"eval_samples_per_second":
|
163 |
-
"eval_steps_per_second":
|
164 |
"step": 1116
|
165 |
},
|
166 |
{
|
167 |
"epoch": 19.0,
|
168 |
-
"eval_loss":
|
169 |
-
"eval_runtime": 3.
|
170 |
-
"eval_samples_per_second":
|
171 |
-
"eval_steps_per_second": 4.
|
172 |
"step": 1178
|
173 |
},
|
174 |
{
|
175 |
"epoch": 20.0,
|
176 |
-
"eval_loss":
|
177 |
-
"eval_runtime": 3.
|
178 |
-
"eval_samples_per_second":
|
179 |
-
"eval_steps_per_second":
|
180 |
"step": 1240
|
181 |
},
|
182 |
{
|
183 |
"epoch": 21.0,
|
184 |
-
"eval_loss":
|
185 |
-
"eval_runtime": 3.
|
186 |
-
"eval_samples_per_second":
|
187 |
-
"eval_steps_per_second": 4.
|
188 |
"step": 1302
|
189 |
},
|
190 |
{
|
191 |
"epoch": 22.0,
|
192 |
-
"eval_loss":
|
193 |
-
"eval_runtime": 3.
|
194 |
-
"eval_samples_per_second":
|
195 |
-
"eval_steps_per_second":
|
196 |
"step": 1364
|
197 |
},
|
198 |
{
|
199 |
"epoch": 23.0,
|
200 |
-
"eval_loss":
|
201 |
-
"eval_runtime": 3.
|
202 |
-
"eval_samples_per_second":
|
203 |
-
"eval_steps_per_second": 4.
|
204 |
"step": 1426
|
205 |
},
|
206 |
{
|
207 |
"epoch": 24.0,
|
208 |
-
"eval_loss":
|
209 |
-
"eval_runtime": 3.
|
210 |
-
"eval_samples_per_second":
|
211 |
-
"eval_steps_per_second":
|
212 |
"step": 1488
|
213 |
},
|
214 |
{
|
215 |
"epoch": 24.19,
|
216 |
-
"learning_rate":
|
217 |
-
"loss":
|
218 |
"step": 1500
|
219 |
},
|
220 |
{
|
221 |
"epoch": 25.0,
|
222 |
-
"eval_loss":
|
223 |
-
"eval_runtime": 3.
|
224 |
-
"eval_samples_per_second":
|
225 |
-
"eval_steps_per_second": 4.
|
226 |
"step": 1550
|
227 |
},
|
228 |
{
|
229 |
"epoch": 26.0,
|
230 |
-
"eval_loss":
|
231 |
-
"eval_runtime": 3.
|
232 |
-
"eval_samples_per_second":
|
233 |
-
"eval_steps_per_second":
|
234 |
"step": 1612
|
235 |
},
|
236 |
{
|
237 |
"epoch": 27.0,
|
238 |
-
"eval_loss":
|
239 |
-
"eval_runtime": 3.
|
240 |
-
"eval_samples_per_second":
|
241 |
-
"eval_steps_per_second": 4.
|
242 |
"step": 1674
|
243 |
},
|
244 |
{
|
245 |
"epoch": 28.0,
|
246 |
-
"eval_loss":
|
247 |
-
"eval_runtime": 3.
|
248 |
-
"eval_samples_per_second":
|
249 |
-
"eval_steps_per_second":
|
250 |
"step": 1736
|
251 |
},
|
252 |
{
|
253 |
"epoch": 29.0,
|
254 |
-
"eval_loss":
|
255 |
-
"eval_runtime": 3.
|
256 |
-
"eval_samples_per_second":
|
257 |
-
"eval_steps_per_second":
|
258 |
"step": 1798
|
259 |
},
|
260 |
{
|
261 |
"epoch": 30.0,
|
262 |
-
"eval_loss":
|
263 |
-
"eval_runtime": 3.
|
264 |
-
"eval_samples_per_second":
|
265 |
-
"eval_steps_per_second": 4.
|
266 |
"step": 1860
|
267 |
},
|
268 |
{
|
269 |
"epoch": 31.0,
|
270 |
-
"eval_loss":
|
271 |
-
"eval_runtime": 3.
|
272 |
-
"eval_samples_per_second":
|
273 |
-
"eval_steps_per_second": 4.
|
274 |
"step": 1922
|
275 |
},
|
276 |
{
|
277 |
"epoch": 32.0,
|
278 |
-
"eval_loss":
|
279 |
-
"eval_runtime": 3.
|
280 |
-
"eval_samples_per_second":
|
281 |
-
"eval_steps_per_second":
|
282 |
"step": 1984
|
283 |
},
|
284 |
{
|
285 |
"epoch": 32.26,
|
286 |
-
"learning_rate":
|
287 |
-
"loss":
|
288 |
"step": 2000
|
289 |
},
|
290 |
{
|
291 |
"epoch": 33.0,
|
292 |
-
"eval_loss":
|
293 |
-
"eval_runtime": 3.
|
294 |
-
"eval_samples_per_second":
|
295 |
-
"eval_steps_per_second": 4.
|
296 |
"step": 2046
|
297 |
},
|
298 |
{
|
299 |
"epoch": 34.0,
|
300 |
-
"eval_loss":
|
301 |
-
"eval_runtime": 3.
|
302 |
-
"eval_samples_per_second":
|
303 |
-
"eval_steps_per_second":
|
304 |
"step": 2108
|
305 |
},
|
306 |
{
|
307 |
"epoch": 35.0,
|
308 |
-
"eval_loss":
|
309 |
-
"eval_runtime": 3.
|
310 |
-
"eval_samples_per_second":
|
311 |
-
"eval_steps_per_second": 4.
|
312 |
"step": 2170
|
313 |
},
|
314 |
{
|
315 |
"epoch": 36.0,
|
316 |
-
"eval_loss":
|
317 |
-
"eval_runtime": 3.
|
318 |
-
"eval_samples_per_second":
|
319 |
-
"eval_steps_per_second":
|
320 |
"step": 2232
|
321 |
},
|
322 |
{
|
323 |
"epoch": 37.0,
|
324 |
-
"eval_loss":
|
325 |
-
"eval_runtime": 3.
|
326 |
-
"eval_samples_per_second":
|
327 |
-
"eval_steps_per_second": 4.
|
328 |
"step": 2294
|
329 |
},
|
330 |
{
|
331 |
"epoch": 38.0,
|
332 |
-
"eval_loss":
|
333 |
-
"eval_runtime": 3.
|
334 |
-
"eval_samples_per_second":
|
335 |
-
"eval_steps_per_second":
|
336 |
"step": 2356
|
337 |
},
|
338 |
{
|
339 |
"epoch": 39.0,
|
340 |
-
"eval_loss":
|
341 |
-
"eval_runtime": 3.
|
342 |
-
"eval_samples_per_second":
|
343 |
-
"eval_steps_per_second": 4.
|
344 |
"step": 2418
|
345 |
},
|
346 |
{
|
347 |
"epoch": 40.0,
|
348 |
-
"eval_loss":
|
349 |
-
"eval_runtime": 3.
|
350 |
-
"eval_samples_per_second":
|
351 |
-
"eval_steps_per_second":
|
352 |
"step": 2480
|
353 |
},
|
354 |
{
|
355 |
"epoch": 40.32,
|
356 |
-
"learning_rate":
|
357 |
-
"loss":
|
358 |
"step": 2500
|
359 |
},
|
360 |
{
|
361 |
"epoch": 41.0,
|
362 |
-
"eval_loss":
|
363 |
-
"eval_runtime": 3.
|
364 |
-
"eval_samples_per_second":
|
365 |
-
"eval_steps_per_second": 4.
|
366 |
"step": 2542
|
367 |
},
|
368 |
{
|
369 |
"epoch": 42.0,
|
370 |
-
"eval_loss":
|
371 |
-
"eval_runtime": 3.
|
372 |
-
"eval_samples_per_second":
|
373 |
-
"eval_steps_per_second":
|
374 |
"step": 2604
|
375 |
},
|
376 |
{
|
377 |
"epoch": 43.0,
|
378 |
-
"eval_loss":
|
379 |
-
"eval_runtime": 3.
|
380 |
-
"eval_samples_per_second":
|
381 |
-
"eval_steps_per_second": 4.
|
382 |
"step": 2666
|
383 |
},
|
384 |
{
|
385 |
"epoch": 44.0,
|
386 |
-
"eval_loss":
|
387 |
-
"eval_runtime": 3.
|
388 |
-
"eval_samples_per_second":
|
389 |
-
"eval_steps_per_second": 4.
|
390 |
"step": 2728
|
391 |
},
|
392 |
{
|
393 |
"epoch": 45.0,
|
394 |
-
"eval_loss":
|
395 |
-
"eval_runtime": 3.
|
396 |
-
"eval_samples_per_second":
|
397 |
-
"eval_steps_per_second":
|
398 |
"step": 2790
|
399 |
},
|
400 |
{
|
401 |
"epoch": 46.0,
|
402 |
-
"eval_loss":
|
403 |
-
"eval_runtime": 3.
|
404 |
-
"eval_samples_per_second":
|
405 |
-
"eval_steps_per_second": 4.
|
406 |
"step": 2852
|
407 |
},
|
408 |
{
|
409 |
"epoch": 47.0,
|
410 |
-
"eval_loss":
|
411 |
-
"eval_runtime": 3.
|
412 |
-
"eval_samples_per_second":
|
413 |
-
"eval_steps_per_second": 4.
|
414 |
"step": 2914
|
415 |
},
|
416 |
{
|
417 |
"epoch": 48.0,
|
418 |
-
"eval_loss":
|
419 |
-
"eval_runtime": 3.
|
420 |
-
"eval_samples_per_second":
|
421 |
-
"eval_steps_per_second": 4.
|
422 |
"step": 2976
|
423 |
},
|
424 |
{
|
425 |
"epoch": 48.39,
|
426 |
-
"learning_rate":
|
427 |
-
"loss":
|
428 |
"step": 3000
|
429 |
},
|
430 |
{
|
431 |
"epoch": 49.0,
|
432 |
-
"eval_loss":
|
433 |
-
"eval_runtime": 3.
|
434 |
-
"eval_samples_per_second":
|
435 |
-
"eval_steps_per_second": 4.
|
436 |
"step": 3038
|
437 |
},
|
438 |
{
|
439 |
"epoch": 50.0,
|
440 |
-
"eval_loss":
|
441 |
-
"eval_runtime": 3.
|
442 |
-
"eval_samples_per_second":
|
443 |
-
"eval_steps_per_second":
|
444 |
"step": 3100
|
445 |
},
|
446 |
{
|
447 |
"epoch": 50.0,
|
448 |
"step": 3100,
|
449 |
"total_flos": 1620261687859200.0,
|
450 |
-
"train_loss":
|
451 |
-
"train_runtime":
|
452 |
-
"train_samples_per_second":
|
453 |
-
"train_steps_per_second": 1.
|
454 |
}
|
455 |
],
|
456 |
"max_steps": 3100,
|
|
|
9 |
"log_history": [
|
10 |
{
|
11 |
"epoch": 1.0,
|
12 |
+
"eval_loss": 39.687320709228516,
|
13 |
+
"eval_runtime": 2.8335,
|
14 |
+
"eval_samples_per_second": 43.056,
|
15 |
+
"eval_steps_per_second": 5.647,
|
16 |
"step": 62
|
17 |
},
|
18 |
{
|
19 |
"epoch": 2.0,
|
20 |
+
"eval_loss": 39.516841888427734,
|
21 |
+
"eval_runtime": 2.9783,
|
22 |
+
"eval_samples_per_second": 40.963,
|
23 |
+
"eval_steps_per_second": 5.372,
|
24 |
"step": 124
|
25 |
},
|
26 |
{
|
27 |
"epoch": 3.0,
|
28 |
+
"eval_loss": 38.932273864746094,
|
29 |
+
"eval_runtime": 3.1472,
|
30 |
+
"eval_samples_per_second": 38.765,
|
31 |
+
"eval_steps_per_second": 5.084,
|
32 |
"step": 186
|
33 |
},
|
34 |
{
|
35 |
"epoch": 4.0,
|
36 |
+
"eval_loss": 37.797115325927734,
|
37 |
+
"eval_runtime": 3.1415,
|
38 |
+
"eval_samples_per_second": 38.835,
|
39 |
+
"eval_steps_per_second": 5.093,
|
40 |
"step": 248
|
41 |
},
|
42 |
{
|
43 |
"epoch": 5.0,
|
44 |
+
"eval_loss": 36.89667510986328,
|
45 |
+
"eval_runtime": 3.1742,
|
46 |
+
"eval_samples_per_second": 38.435,
|
47 |
+
"eval_steps_per_second": 5.041,
|
48 |
"step": 310
|
49 |
},
|
50 |
{
|
51 |
"epoch": 6.0,
|
52 |
+
"eval_loss": 35.38439178466797,
|
53 |
+
"eval_runtime": 3.1525,
|
54 |
+
"eval_samples_per_second": 38.699,
|
55 |
+
"eval_steps_per_second": 5.075,
|
56 |
"step": 372
|
57 |
},
|
58 |
{
|
59 |
"epoch": 7.0,
|
60 |
+
"eval_loss": 34.643463134765625,
|
61 |
+
"eval_runtime": 3.1776,
|
62 |
+
"eval_samples_per_second": 38.394,
|
63 |
+
"eval_steps_per_second": 5.035,
|
64 |
"step": 434
|
65 |
},
|
66 |
{
|
67 |
"epoch": 8.0,
|
68 |
+
"eval_loss": 33.6444206237793,
|
69 |
+
"eval_runtime": 3.1827,
|
70 |
+
"eval_samples_per_second": 38.332,
|
71 |
+
"eval_steps_per_second": 5.027,
|
72 |
"step": 496
|
73 |
},
|
74 |
{
|
75 |
"epoch": 8.06,
|
76 |
+
"learning_rate": 5.870967741935484e-08,
|
77 |
+
"loss": 39.4735,
|
78 |
"step": 500
|
79 |
},
|
80 |
{
|
81 |
"epoch": 9.0,
|
82 |
+
"eval_loss": 32.891536712646484,
|
83 |
+
"eval_runtime": 3.2424,
|
84 |
+
"eval_samples_per_second": 37.627,
|
85 |
+
"eval_steps_per_second": 4.935,
|
86 |
"step": 558
|
87 |
},
|
88 |
{
|
89 |
"epoch": 10.0,
|
90 |
+
"eval_loss": 31.138931274414062,
|
91 |
+
"eval_runtime": 3.1172,
|
92 |
+
"eval_samples_per_second": 39.138,
|
93 |
+
"eval_steps_per_second": 5.133,
|
94 |
"step": 620
|
95 |
},
|
96 |
{
|
97 |
"epoch": 11.0,
|
98 |
+
"eval_loss": 30.020915985107422,
|
99 |
+
"eval_runtime": 3.1861,
|
100 |
+
"eval_samples_per_second": 38.291,
|
101 |
+
"eval_steps_per_second": 5.022,
|
102 |
"step": 682
|
103 |
},
|
104 |
{
|
105 |
"epoch": 12.0,
|
106 |
+
"eval_loss": 28.798078536987305,
|
107 |
+
"eval_runtime": 3.158,
|
108 |
+
"eval_samples_per_second": 38.632,
|
109 |
+
"eval_steps_per_second": 5.066,
|
110 |
"step": 744
|
111 |
},
|
112 |
{
|
113 |
"epoch": 13.0,
|
114 |
+
"eval_loss": 27.6728458404541,
|
115 |
+
"eval_runtime": 3.1725,
|
116 |
+
"eval_samples_per_second": 38.456,
|
117 |
+
"eval_steps_per_second": 5.043,
|
118 |
"step": 806
|
119 |
},
|
120 |
{
|
121 |
"epoch": 14.0,
|
122 |
+
"eval_loss": 26.61111068725586,
|
123 |
+
"eval_runtime": 3.1711,
|
124 |
+
"eval_samples_per_second": 38.472,
|
125 |
+
"eval_steps_per_second": 5.046,
|
126 |
"step": 868
|
127 |
},
|
128 |
{
|
129 |
"epoch": 15.0,
|
130 |
+
"eval_loss": 25.55615234375,
|
131 |
+
"eval_runtime": 3.1645,
|
132 |
+
"eval_samples_per_second": 38.553,
|
133 |
+
"eval_steps_per_second": 5.056,
|
134 |
"step": 930
|
135 |
},
|
136 |
{
|
137 |
"epoch": 16.0,
|
138 |
+
"eval_loss": 24.428285598754883,
|
139 |
+
"eval_runtime": 3.1894,
|
140 |
+
"eval_samples_per_second": 38.251,
|
141 |
+
"eval_steps_per_second": 5.017,
|
142 |
"step": 992
|
143 |
},
|
144 |
{
|
145 |
"epoch": 16.13,
|
146 |
+
"learning_rate": 4.741935483870968e-08,
|
147 |
+
"loss": 29.1909,
|
148 |
"step": 1000
|
149 |
},
|
150 |
{
|
151 |
"epoch": 17.0,
|
152 |
+
"eval_loss": 23.227224349975586,
|
153 |
+
"eval_runtime": 3.3011,
|
154 |
+
"eval_samples_per_second": 36.957,
|
155 |
+
"eval_steps_per_second": 4.847,
|
156 |
"step": 1054
|
157 |
},
|
158 |
{
|
159 |
"epoch": 18.0,
|
160 |
+
"eval_loss": 22.856584548950195,
|
161 |
+
"eval_runtime": 3.1351,
|
162 |
+
"eval_samples_per_second": 38.914,
|
163 |
+
"eval_steps_per_second": 5.103,
|
164 |
"step": 1116
|
165 |
},
|
166 |
{
|
167 |
"epoch": 19.0,
|
168 |
+
"eval_loss": 22.174100875854492,
|
169 |
+
"eval_runtime": 3.2113,
|
170 |
+
"eval_samples_per_second": 37.99,
|
171 |
+
"eval_steps_per_second": 4.982,
|
172 |
"step": 1178
|
173 |
},
|
174 |
{
|
175 |
"epoch": 20.0,
|
176 |
+
"eval_loss": 21.41921615600586,
|
177 |
+
"eval_runtime": 3.1698,
|
178 |
+
"eval_samples_per_second": 38.488,
|
179 |
+
"eval_steps_per_second": 5.048,
|
180 |
"step": 1240
|
181 |
},
|
182 |
{
|
183 |
"epoch": 21.0,
|
184 |
+
"eval_loss": 20.49114227294922,
|
185 |
+
"eval_runtime": 3.202,
|
186 |
+
"eval_samples_per_second": 38.101,
|
187 |
+
"eval_steps_per_second": 4.997,
|
188 |
"step": 1302
|
189 |
},
|
190 |
{
|
191 |
"epoch": 22.0,
|
192 |
+
"eval_loss": 20.245670318603516,
|
193 |
+
"eval_runtime": 3.1786,
|
194 |
+
"eval_samples_per_second": 38.382,
|
195 |
+
"eval_steps_per_second": 5.034,
|
196 |
"step": 1364
|
197 |
},
|
198 |
{
|
199 |
"epoch": 23.0,
|
200 |
+
"eval_loss": 19.58565902709961,
|
201 |
+
"eval_runtime": 3.205,
|
202 |
+
"eval_samples_per_second": 38.065,
|
203 |
+
"eval_steps_per_second": 4.992,
|
204 |
"step": 1426
|
205 |
},
|
206 |
{
|
207 |
"epoch": 24.0,
|
208 |
+
"eval_loss": 19.287837982177734,
|
209 |
+
"eval_runtime": 3.1902,
|
210 |
+
"eval_samples_per_second": 38.242,
|
211 |
+
"eval_steps_per_second": 5.015,
|
212 |
"step": 1488
|
213 |
},
|
214 |
{
|
215 |
"epoch": 24.19,
|
216 |
+
"learning_rate": 3.612903225806452e-08,
|
217 |
+
"loss": 21.7917,
|
218 |
"step": 1500
|
219 |
},
|
220 |
{
|
221 |
"epoch": 25.0,
|
222 |
+
"eval_loss": 18.806514739990234,
|
223 |
+
"eval_runtime": 3.3778,
|
224 |
+
"eval_samples_per_second": 36.118,
|
225 |
+
"eval_steps_per_second": 4.737,
|
226 |
"step": 1550
|
227 |
},
|
228 |
{
|
229 |
"epoch": 26.0,
|
230 |
+
"eval_loss": 18.336509704589844,
|
231 |
+
"eval_runtime": 3.1436,
|
232 |
+
"eval_samples_per_second": 38.81,
|
233 |
+
"eval_steps_per_second": 5.09,
|
234 |
"step": 1612
|
235 |
},
|
236 |
{
|
237 |
"epoch": 27.0,
|
238 |
+
"eval_loss": 18.16900062561035,
|
239 |
+
"eval_runtime": 3.2259,
|
240 |
+
"eval_samples_per_second": 37.819,
|
241 |
+
"eval_steps_per_second": 4.96,
|
242 |
"step": 1674
|
243 |
},
|
244 |
{
|
245 |
"epoch": 28.0,
|
246 |
+
"eval_loss": 17.55986213684082,
|
247 |
+
"eval_runtime": 3.1926,
|
248 |
+
"eval_samples_per_second": 38.213,
|
249 |
+
"eval_steps_per_second": 5.012,
|
250 |
"step": 1736
|
251 |
},
|
252 |
{
|
253 |
"epoch": 29.0,
|
254 |
+
"eval_loss": 17.429519653320312,
|
255 |
+
"eval_runtime": 3.19,
|
256 |
+
"eval_samples_per_second": 38.245,
|
257 |
+
"eval_steps_per_second": 5.016,
|
258 |
"step": 1798
|
259 |
},
|
260 |
{
|
261 |
"epoch": 30.0,
|
262 |
+
"eval_loss": 17.191831588745117,
|
263 |
+
"eval_runtime": 3.2186,
|
264 |
+
"eval_samples_per_second": 37.905,
|
265 |
+
"eval_steps_per_second": 4.971,
|
266 |
"step": 1860
|
267 |
},
|
268 |
{
|
269 |
"epoch": 31.0,
|
270 |
+
"eval_loss": 16.991268157958984,
|
271 |
+
"eval_runtime": 3.2004,
|
272 |
+
"eval_samples_per_second": 38.12,
|
273 |
+
"eval_steps_per_second": 4.999,
|
274 |
"step": 1922
|
275 |
},
|
276 |
{
|
277 |
"epoch": 32.0,
|
278 |
+
"eval_loss": 16.737815856933594,
|
279 |
+
"eval_runtime": 3.2002,
|
280 |
+
"eval_samples_per_second": 38.122,
|
281 |
+
"eval_steps_per_second": 5.0,
|
282 |
"step": 1984
|
283 |
},
|
284 |
{
|
285 |
"epoch": 32.26,
|
286 |
+
"learning_rate": 2.483870967741936e-08,
|
287 |
+
"loss": 18.0017,
|
288 |
"step": 2000
|
289 |
},
|
290 |
{
|
291 |
"epoch": 33.0,
|
292 |
+
"eval_loss": 16.585847854614258,
|
293 |
+
"eval_runtime": 3.3866,
|
294 |
+
"eval_samples_per_second": 36.024,
|
295 |
+
"eval_steps_per_second": 4.725,
|
296 |
"step": 2046
|
297 |
},
|
298 |
{
|
299 |
"epoch": 34.0,
|
300 |
+
"eval_loss": 16.41578483581543,
|
301 |
+
"eval_runtime": 3.1469,
|
302 |
+
"eval_samples_per_second": 38.769,
|
303 |
+
"eval_steps_per_second": 5.084,
|
304 |
"step": 2108
|
305 |
},
|
306 |
{
|
307 |
"epoch": 35.0,
|
308 |
+
"eval_loss": 16.26023292541504,
|
309 |
+
"eval_runtime": 3.2254,
|
310 |
+
"eval_samples_per_second": 37.825,
|
311 |
+
"eval_steps_per_second": 4.961,
|
312 |
"step": 2170
|
313 |
},
|
314 |
{
|
315 |
"epoch": 36.0,
|
316 |
+
"eval_loss": 16.033435821533203,
|
317 |
+
"eval_runtime": 3.1846,
|
318 |
+
"eval_samples_per_second": 38.309,
|
319 |
+
"eval_steps_per_second": 5.024,
|
320 |
"step": 2232
|
321 |
},
|
322 |
{
|
323 |
"epoch": 37.0,
|
324 |
+
"eval_loss": 16.080913543701172,
|
325 |
+
"eval_runtime": 3.2091,
|
326 |
+
"eval_samples_per_second": 38.017,
|
327 |
+
"eval_steps_per_second": 4.986,
|
328 |
"step": 2294
|
329 |
},
|
330 |
{
|
331 |
"epoch": 38.0,
|
332 |
+
"eval_loss": 15.7175931930542,
|
333 |
+
"eval_runtime": 3.1978,
|
334 |
+
"eval_samples_per_second": 38.151,
|
335 |
+
"eval_steps_per_second": 5.003,
|
336 |
"step": 2356
|
337 |
},
|
338 |
{
|
339 |
"epoch": 39.0,
|
340 |
+
"eval_loss": 15.824816703796387,
|
341 |
+
"eval_runtime": 3.2153,
|
342 |
+
"eval_samples_per_second": 37.944,
|
343 |
+
"eval_steps_per_second": 4.976,
|
344 |
"step": 2418
|
345 |
},
|
346 |
{
|
347 |
"epoch": 40.0,
|
348 |
+
"eval_loss": 15.852823257446289,
|
349 |
+
"eval_runtime": 3.1937,
|
350 |
+
"eval_samples_per_second": 38.2,
|
351 |
+
"eval_steps_per_second": 5.01,
|
352 |
"step": 2480
|
353 |
},
|
354 |
{
|
355 |
"epoch": 40.32,
|
356 |
+
"learning_rate": 1.3548387096774193e-08,
|
357 |
+
"loss": 16.124,
|
358 |
"step": 2500
|
359 |
},
|
360 |
{
|
361 |
"epoch": 41.0,
|
362 |
+
"eval_loss": 15.77718734741211,
|
363 |
+
"eval_runtime": 3.377,
|
364 |
+
"eval_samples_per_second": 36.127,
|
365 |
+
"eval_steps_per_second": 4.738,
|
366 |
"step": 2542
|
367 |
},
|
368 |
{
|
369 |
"epoch": 42.0,
|
370 |
+
"eval_loss": 15.797329902648926,
|
371 |
+
"eval_runtime": 3.1524,
|
372 |
+
"eval_samples_per_second": 38.701,
|
373 |
+
"eval_steps_per_second": 5.075,
|
374 |
"step": 2604
|
375 |
},
|
376 |
{
|
377 |
"epoch": 43.0,
|
378 |
+
"eval_loss": 15.45718002319336,
|
379 |
+
"eval_runtime": 3.2372,
|
380 |
+
"eval_samples_per_second": 37.686,
|
381 |
+
"eval_steps_per_second": 4.942,
|
382 |
"step": 2666
|
383 |
},
|
384 |
{
|
385 |
"epoch": 44.0,
|
386 |
+
"eval_loss": 15.622541427612305,
|
387 |
+
"eval_runtime": 3.2026,
|
388 |
+
"eval_samples_per_second": 38.094,
|
389 |
+
"eval_steps_per_second": 4.996,
|
390 |
"step": 2728
|
391 |
},
|
392 |
{
|
393 |
"epoch": 45.0,
|
394 |
+
"eval_loss": 15.430057525634766,
|
395 |
+
"eval_runtime": 3.1974,
|
396 |
+
"eval_samples_per_second": 38.156,
|
397 |
+
"eval_steps_per_second": 5.004,
|
398 |
"step": 2790
|
399 |
},
|
400 |
{
|
401 |
"epoch": 46.0,
|
402 |
+
"eval_loss": 15.735359191894531,
|
403 |
+
"eval_runtime": 3.2169,
|
404 |
+
"eval_samples_per_second": 37.925,
|
405 |
+
"eval_steps_per_second": 4.974,
|
406 |
"step": 2852
|
407 |
},
|
408 |
{
|
409 |
"epoch": 47.0,
|
410 |
+
"eval_loss": 15.529631614685059,
|
411 |
+
"eval_runtime": 3.2177,
|
412 |
+
"eval_samples_per_second": 37.916,
|
413 |
+
"eval_steps_per_second": 4.973,
|
414 |
"step": 2914
|
415 |
},
|
416 |
{
|
417 |
"epoch": 48.0,
|
418 |
+
"eval_loss": 15.421342849731445,
|
419 |
+
"eval_runtime": 3.225,
|
420 |
+
"eval_samples_per_second": 37.829,
|
421 |
+
"eval_steps_per_second": 4.961,
|
422 |
"step": 2976
|
423 |
},
|
424 |
{
|
425 |
"epoch": 48.39,
|
426 |
+
"learning_rate": 2.2580645161290324e-09,
|
427 |
+
"loss": 15.3787,
|
428 |
"step": 3000
|
429 |
},
|
430 |
{
|
431 |
"epoch": 49.0,
|
432 |
+
"eval_loss": 15.512802124023438,
|
433 |
+
"eval_runtime": 3.3809,
|
434 |
+
"eval_samples_per_second": 36.085,
|
435 |
+
"eval_steps_per_second": 4.732,
|
436 |
"step": 3038
|
437 |
},
|
438 |
{
|
439 |
"epoch": 50.0,
|
440 |
+
"eval_loss": 15.596056938171387,
|
441 |
+
"eval_runtime": 3.151,
|
442 |
+
"eval_samples_per_second": 38.718,
|
443 |
+
"eval_steps_per_second": 5.078,
|
444 |
"step": 3100
|
445 |
},
|
446 |
{
|
447 |
"epoch": 50.0,
|
448 |
"step": 3100,
|
449 |
"total_flos": 1620261687859200.0,
|
450 |
+
"train_loss": 23.064505024571574,
|
451 |
+
"train_runtime": 2425.9558,
|
452 |
+
"train_samples_per_second": 10.12,
|
453 |
+
"train_steps_per_second": 1.278
|
454 |
}
|
455 |
],
|
456 |
"max_steps": 3100,
|