JEdward7777
commited on
Commit
•
ea59389
1
Parent(s):
fa20d62
End of training
Browse files- all_results.json +11 -11
- eval_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +331 -319
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
-
"epoch": 39.
|
3 |
-
"eval_accuracy":
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime":
|
6 |
-
"eval_samples_per_second": 5.
|
7 |
-
"eval_steps_per_second": 0.
|
8 |
-
"total_flos":
|
9 |
-
"train_loss":
|
10 |
-
"train_runtime":
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second": 0.
|
13 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 39.8,
|
3 |
+
"eval_accuracy": 0.9714285714285714,
|
4 |
+
"eval_loss": 0.1382756382226944,
|
5 |
+
"eval_runtime": 6.6641,
|
6 |
+
"eval_samples_per_second": 5.252,
|
7 |
+
"eval_steps_per_second": 0.3,
|
8 |
+
"total_flos": 3.088453228308726e+17,
|
9 |
+
"train_loss": 0.15159874260425568,
|
10 |
+
"train_runtime": 6474.8036,
|
11 |
+
"train_samples_per_second": 1.927,
|
12 |
+
"train_steps_per_second": 0.012
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch": 39.
|
3 |
-
"eval_accuracy":
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime":
|
6 |
-
"eval_samples_per_second": 5.
|
7 |
-
"eval_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 39.8,
|
3 |
+
"eval_accuracy": 0.9714285714285714,
|
4 |
+
"eval_loss": 0.1382756382226944,
|
5 |
+
"eval_runtime": 6.6641,
|
6 |
+
"eval_samples_per_second": 5.252,
|
7 |
+
"eval_steps_per_second": 0.3
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch": 39.
|
3 |
-
"total_flos":
|
4 |
-
"train_loss":
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 39.8,
|
3 |
+
"total_flos": 3.088453228308726e+17,
|
4 |
+
"train_loss": 0.15159874260425568,
|
5 |
+
"train_runtime": 6474.8036,
|
6 |
+
"train_samples_per_second": 1.927,
|
7 |
+
"train_steps_per_second": 0.012
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,397 +1,409 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "delivery_truck_classification
|
4 |
-
"epoch": 39.
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
-
"epoch": 0.
|
12 |
-
"eval_accuracy": 0.
|
13 |
-
"eval_loss":
|
14 |
-
"eval_runtime":
|
15 |
-
"eval_samples_per_second":
|
16 |
-
"eval_steps_per_second": 0.
|
17 |
-
"step": 1
|
18 |
-
},
|
19 |
-
{
|
20 |
-
"epoch": 1.57,
|
21 |
-
"eval_accuracy": 0.09090909090909091,
|
22 |
-
"eval_loss": 1.9021785259246826,
|
23 |
-
"eval_runtime": 1.7599,
|
24 |
-
"eval_samples_per_second": 12.501,
|
25 |
-
"eval_steps_per_second": 0.568,
|
26 |
"step": 2
|
27 |
},
|
28 |
{
|
29 |
-
"epoch":
|
30 |
-
"eval_accuracy": 0.
|
31 |
-
"eval_loss":
|
32 |
-
"eval_runtime":
|
33 |
-
"eval_samples_per_second":
|
34 |
-
"eval_steps_per_second": 0.
|
35 |
-
"step": 3
|
36 |
-
},
|
37 |
-
{
|
38 |
-
"epoch": 3.57,
|
39 |
-
"eval_accuracy": 0.36363636363636365,
|
40 |
-
"eval_loss": 1.646571397781372,
|
41 |
-
"eval_runtime": 1.8048,
|
42 |
-
"eval_samples_per_second": 12.19,
|
43 |
-
"eval_steps_per_second": 0.554,
|
44 |
"step": 4
|
45 |
},
|
46 |
{
|
47 |
-
"epoch":
|
48 |
-
"eval_accuracy": 0.
|
49 |
-
"eval_loss":
|
50 |
-
"eval_runtime":
|
51 |
-
"eval_samples_per_second":
|
52 |
-
"eval_steps_per_second": 0.
|
53 |
-
"step": 5
|
54 |
-
},
|
55 |
-
{
|
56 |
-
"epoch": 5.57,
|
57 |
-
"eval_accuracy": 0.5454545454545454,
|
58 |
-
"eval_loss": 1.4261207580566406,
|
59 |
-
"eval_runtime": 1.6096,
|
60 |
-
"eval_samples_per_second": 13.668,
|
61 |
-
"eval_steps_per_second": 0.621,
|
62 |
"step": 6
|
63 |
},
|
64 |
{
|
65 |
-
"epoch":
|
66 |
-
"eval_accuracy": 0.
|
67 |
-
"eval_loss":
|
68 |
-
"eval_runtime":
|
69 |
-
"eval_samples_per_second":
|
70 |
-
"eval_steps_per_second": 0.
|
71 |
-
"step": 7
|
72 |
-
},
|
73 |
-
{
|
74 |
-
"epoch": 7.57,
|
75 |
-
"eval_accuracy": 0.6363636363636364,
|
76 |
-
"eval_loss": 1.2797943353652954,
|
77 |
-
"eval_runtime": 1.8043,
|
78 |
-
"eval_samples_per_second": 12.193,
|
79 |
-
"eval_steps_per_second": 0.554,
|
80 |
"step": 8
|
81 |
},
|
82 |
{
|
83 |
-
"epoch": 8
|
84 |
-
"eval_accuracy": 0.
|
85 |
-
"eval_loss":
|
86 |
-
"eval_runtime":
|
87 |
-
"eval_samples_per_second":
|
88 |
-
"eval_steps_per_second": 0.
|
89 |
-
"step": 9
|
90 |
-
},
|
91 |
-
{
|
92 |
-
"epoch": 9.57,
|
93 |
-
"eval_accuracy": 0.6363636363636364,
|
94 |
-
"eval_loss": 1.1594452857971191,
|
95 |
-
"eval_runtime": 1.803,
|
96 |
-
"eval_samples_per_second": 12.202,
|
97 |
-
"eval_steps_per_second": 0.555,
|
98 |
"step": 10
|
99 |
},
|
100 |
{
|
101 |
-
"epoch":
|
102 |
-
"eval_accuracy": 0.
|
103 |
-
"eval_loss":
|
104 |
-
"eval_runtime":
|
105 |
-
"eval_samples_per_second":
|
106 |
-
"eval_steps_per_second": 0.
|
107 |
-
"step": 11
|
108 |
-
},
|
109 |
-
{
|
110 |
-
"epoch": 11.57,
|
111 |
-
"eval_accuracy": 0.6818181818181818,
|
112 |
-
"eval_loss": 1.07809579372406,
|
113 |
-
"eval_runtime": 1.6416,
|
114 |
-
"eval_samples_per_second": 13.401,
|
115 |
-
"eval_steps_per_second": 0.609,
|
116 |
"step": 12
|
117 |
},
|
118 |
{
|
119 |
-
"epoch":
|
120 |
-
"eval_accuracy": 0.
|
121 |
-
"eval_loss":
|
122 |
-
"eval_runtime":
|
123 |
-
"eval_samples_per_second":
|
124 |
-
"eval_steps_per_second": 0.
|
125 |
-
"step": 13
|
126 |
-
},
|
127 |
-
{
|
128 |
-
"epoch": 13.57,
|
129 |
-
"eval_accuracy": 0.6818181818181818,
|
130 |
-
"eval_loss": 0.9623335003852844,
|
131 |
-
"eval_runtime": 1.8244,
|
132 |
-
"eval_samples_per_second": 12.059,
|
133 |
-
"eval_steps_per_second": 0.548,
|
134 |
"step": 14
|
135 |
},
|
136 |
{
|
137 |
-
"epoch":
|
138 |
-
"eval_accuracy": 0.
|
139 |
-
"eval_loss": 0.
|
140 |
-
"eval_runtime":
|
141 |
-
"eval_samples_per_second":
|
142 |
-
"eval_steps_per_second": 0.
|
143 |
-
"step": 15
|
144 |
-
},
|
145 |
-
{
|
146 |
-
"epoch": 15.57,
|
147 |
-
"eval_accuracy": 0.7272727272727273,
|
148 |
-
"eval_loss": 0.8217518329620361,
|
149 |
-
"eval_runtime": 1.8059,
|
150 |
-
"eval_samples_per_second": 12.182,
|
151 |
-
"eval_steps_per_second": 0.554,
|
152 |
"step": 16
|
153 |
},
|
154 |
{
|
155 |
-
"epoch":
|
156 |
-
"eval_accuracy": 0.
|
157 |
-
"eval_loss": 0.
|
158 |
-
"eval_runtime":
|
159 |
-
"eval_samples_per_second":
|
160 |
-
"eval_steps_per_second": 0.
|
161 |
-
"step": 17
|
162 |
-
},
|
163 |
-
{
|
164 |
-
"epoch": 17.57,
|
165 |
-
"eval_accuracy": 0.8181818181818182,
|
166 |
-
"eval_loss": 0.6525326371192932,
|
167 |
-
"eval_runtime": 1.8033,
|
168 |
-
"eval_samples_per_second": 12.2,
|
169 |
-
"eval_steps_per_second": 0.555,
|
170 |
"step": 18
|
171 |
},
|
172 |
{
|
173 |
-
"epoch":
|
174 |
-
"
|
175 |
-
"
|
176 |
-
"eval_runtime": 1.7903,
|
177 |
-
"eval_samples_per_second": 12.288,
|
178 |
-
"eval_steps_per_second": 0.559,
|
179 |
-
"step": 19
|
180 |
-
},
|
181 |
-
{
|
182 |
-
"epoch": 19.57,
|
183 |
-
"learning_rate": 2.777777777777778e-05,
|
184 |
-
"loss": 1.9399,
|
185 |
"step": 20
|
186 |
},
|
187 |
{
|
188 |
-
"epoch":
|
189 |
-
"eval_accuracy": 0.
|
190 |
-
"eval_loss": 0.
|
191 |
-
"eval_runtime":
|
192 |
-
"eval_samples_per_second":
|
193 |
-
"eval_steps_per_second": 0.
|
194 |
"step": 20
|
195 |
},
|
196 |
{
|
197 |
-
"epoch":
|
198 |
-
"eval_accuracy": 0.
|
199 |
-
"eval_loss": 0.
|
200 |
-
"eval_runtime":
|
201 |
-
"eval_samples_per_second":
|
202 |
-
"eval_steps_per_second": 0.
|
203 |
-
"step": 21
|
204 |
-
},
|
205 |
-
{
|
206 |
-
"epoch": 21.57,
|
207 |
-
"eval_accuracy": 0.9090909090909091,
|
208 |
-
"eval_loss": 0.4493587613105774,
|
209 |
-
"eval_runtime": 1.6389,
|
210 |
-
"eval_samples_per_second": 13.424,
|
211 |
-
"eval_steps_per_second": 0.61,
|
212 |
"step": 22
|
213 |
},
|
214 |
{
|
215 |
-
"epoch":
|
216 |
-
"eval_accuracy": 0.
|
217 |
-
"eval_loss": 0.
|
218 |
-
"eval_runtime":
|
219 |
-
"eval_samples_per_second":
|
220 |
-
"eval_steps_per_second": 0.
|
221 |
-
"step": 23
|
222 |
-
},
|
223 |
-
{
|
224 |
-
"epoch": 23.57,
|
225 |
-
"eval_accuracy": 0.8636363636363636,
|
226 |
-
"eval_loss": 0.4357987344264984,
|
227 |
-
"eval_runtime": 1.7864,
|
228 |
-
"eval_samples_per_second": 12.315,
|
229 |
-
"eval_steps_per_second": 0.56,
|
230 |
"step": 24
|
231 |
},
|
232 |
{
|
233 |
-
"epoch":
|
234 |
-
"eval_accuracy": 0.
|
235 |
-
"eval_loss": 0.
|
236 |
-
"eval_runtime":
|
237 |
-
"eval_samples_per_second":
|
238 |
-
"eval_steps_per_second": 0.
|
239 |
-
"step": 25
|
240 |
-
},
|
241 |
-
{
|
242 |
-
"epoch": 25.57,
|
243 |
-
"eval_accuracy": 0.8636363636363636,
|
244 |
-
"eval_loss": 0.4099968671798706,
|
245 |
-
"eval_runtime": 1.7489,
|
246 |
-
"eval_samples_per_second": 12.579,
|
247 |
-
"eval_steps_per_second": 0.572,
|
248 |
"step": 26
|
249 |
},
|
250 |
{
|
251 |
-
"epoch":
|
252 |
-
"eval_accuracy": 0.
|
253 |
-
"eval_loss": 0.
|
254 |
-
"eval_runtime":
|
255 |
-
"eval_samples_per_second":
|
256 |
-
"eval_steps_per_second": 0.
|
257 |
-
"step": 27
|
258 |
-
},
|
259 |
-
{
|
260 |
-
"epoch": 27.57,
|
261 |
-
"eval_accuracy": 0.8181818181818182,
|
262 |
-
"eval_loss": 0.3818438649177551,
|
263 |
-
"eval_runtime": 1.8073,
|
264 |
-
"eval_samples_per_second": 12.173,
|
265 |
-
"eval_steps_per_second": 0.553,
|
266 |
"step": 28
|
267 |
},
|
268 |
{
|
269 |
-
"epoch":
|
270 |
-
"eval_accuracy": 0.
|
271 |
-
"eval_loss": 0.
|
272 |
-
"eval_runtime":
|
273 |
-
"eval_samples_per_second":
|
274 |
-
"eval_steps_per_second": 0.
|
275 |
-
"step": 29
|
276 |
-
},
|
277 |
-
{
|
278 |
-
"epoch": 29.57,
|
279 |
-
"eval_accuracy": 0.8636363636363636,
|
280 |
-
"eval_loss": 0.35147225856781006,
|
281 |
-
"eval_runtime": 1.8109,
|
282 |
-
"eval_samples_per_second": 12.149,
|
283 |
-
"eval_steps_per_second": 0.552,
|
284 |
"step": 30
|
285 |
},
|
286 |
{
|
287 |
-
"epoch":
|
288 |
-
"eval_accuracy": 0.
|
289 |
-
"eval_loss": 0.
|
290 |
-
"eval_runtime":
|
291 |
-
"eval_samples_per_second":
|
292 |
-
"eval_steps_per_second": 0.
|
293 |
-
"step": 31
|
294 |
-
},
|
295 |
-
{
|
296 |
-
"epoch": 31.57,
|
297 |
-
"eval_accuracy": 0.8636363636363636,
|
298 |
-
"eval_loss": 0.32043513655662537,
|
299 |
-
"eval_runtime": 1.599,
|
300 |
-
"eval_samples_per_second": 13.759,
|
301 |
-
"eval_steps_per_second": 0.625,
|
302 |
"step": 32
|
303 |
},
|
304 |
{
|
305 |
-
"epoch":
|
306 |
-
"eval_accuracy": 0.
|
307 |
-
"eval_loss": 0.
|
308 |
-
"eval_runtime":
|
309 |
-
"eval_samples_per_second":
|
310 |
-
"eval_steps_per_second": 0.
|
311 |
-
"step": 33
|
312 |
-
},
|
313 |
-
{
|
314 |
-
"epoch": 33.57,
|
315 |
-
"eval_accuracy": 0.8636363636363636,
|
316 |
-
"eval_loss": 0.2948474586009979,
|
317 |
-
"eval_runtime": 1.7885,
|
318 |
-
"eval_samples_per_second": 12.301,
|
319 |
-
"eval_steps_per_second": 0.559,
|
320 |
"step": 34
|
321 |
},
|
322 |
{
|
323 |
-
"epoch":
|
324 |
-
"eval_accuracy": 0.
|
325 |
-
"eval_loss": 0.
|
326 |
-
"eval_runtime":
|
327 |
-
"eval_samples_per_second":
|
328 |
-
"eval_steps_per_second": 0.
|
329 |
-
"step": 35
|
330 |
-
},
|
331 |
-
{
|
332 |
-
"epoch": 35.57,
|
333 |
-
"eval_accuracy": 0.8636363636363636,
|
334 |
-
"eval_loss": 0.2748388946056366,
|
335 |
-
"eval_runtime": 1.8319,
|
336 |
-
"eval_samples_per_second": 12.009,
|
337 |
-
"eval_steps_per_second": 0.546,
|
338 |
"step": 36
|
339 |
},
|
340 |
{
|
341 |
-
"epoch":
|
342 |
-
"eval_accuracy": 0.
|
343 |
-
"eval_loss": 0.
|
344 |
-
"eval_runtime":
|
345 |
-
"eval_samples_per_second":
|
346 |
-
"eval_steps_per_second": 0.
|
347 |
-
"step": 37
|
348 |
-
},
|
349 |
-
{
|
350 |
-
"epoch": 37.57,
|
351 |
-
"eval_accuracy": 0.8636363636363636,
|
352 |
-
"eval_loss": 0.26424679160118103,
|
353 |
-
"eval_runtime": 1.8361,
|
354 |
-
"eval_samples_per_second": 11.982,
|
355 |
-
"eval_steps_per_second": 0.545,
|
356 |
"step": 38
|
357 |
},
|
358 |
{
|
359 |
-
"epoch":
|
360 |
-
"
|
361 |
-
"
|
362 |
-
"eval_runtime": 1.7769,
|
363 |
-
"eval_samples_per_second": 12.381,
|
364 |
-
"eval_steps_per_second": 0.563,
|
365 |
-
"step": 39
|
366 |
-
},
|
367 |
-
{
|
368 |
-
"epoch": 39.57,
|
369 |
-
"learning_rate": 0.0,
|
370 |
-
"loss": 0.728,
|
371 |
"step": 40
|
372 |
},
|
373 |
{
|
374 |
-
"epoch":
|
375 |
-
"eval_accuracy": 0.
|
376 |
-
"eval_loss": 0.
|
377 |
-
"eval_runtime":
|
378 |
-
"eval_samples_per_second":
|
379 |
-
"eval_steps_per_second": 0.
|
380 |
"step": 40
|
381 |
},
|
382 |
{
|
383 |
-
"epoch":
|
384 |
-
"
|
385 |
-
"
|
386 |
-
"
|
387 |
-
"
|
388 |
-
"
|
389 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
390 |
}
|
391 |
],
|
392 |
-
"max_steps":
|
393 |
"num_train_epochs": 40,
|
394 |
-
"total_flos":
|
395 |
"trial_name": null,
|
396 |
"trial_params": null
|
397 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.9714285714285714,
|
3 |
+
"best_model_checkpoint": "delivery_truck_classification\\checkpoint-4",
|
4 |
+
"epoch": 39.8,
|
5 |
+
"global_step": 80,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
+
"epoch": 0.8,
|
12 |
+
"eval_accuracy": 0.9428571428571428,
|
13 |
+
"eval_loss": 0.19194960594177246,
|
14 |
+
"eval_runtime": 8.8411,
|
15 |
+
"eval_samples_per_second": 3.959,
|
16 |
+
"eval_steps_per_second": 0.226,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
"step": 2
|
18 |
},
|
19 |
{
|
20 |
+
"epoch": 1.8,
|
21 |
+
"eval_accuracy": 0.9714285714285714,
|
22 |
+
"eval_loss": 0.1382756382226944,
|
23 |
+
"eval_runtime": 7.2367,
|
24 |
+
"eval_samples_per_second": 4.836,
|
25 |
+
"eval_steps_per_second": 0.276,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
"step": 4
|
27 |
},
|
28 |
{
|
29 |
+
"epoch": 2.8,
|
30 |
+
"eval_accuracy": 0.9142857142857143,
|
31 |
+
"eval_loss": 0.19303296506404877,
|
32 |
+
"eval_runtime": 7.0677,
|
33 |
+
"eval_samples_per_second": 4.952,
|
34 |
+
"eval_steps_per_second": 0.283,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
"step": 6
|
36 |
},
|
37 |
{
|
38 |
+
"epoch": 3.8,
|
39 |
+
"eval_accuracy": 0.9714285714285714,
|
40 |
+
"eval_loss": 0.1462627798318863,
|
41 |
+
"eval_runtime": 7.1631,
|
42 |
+
"eval_samples_per_second": 4.886,
|
43 |
+
"eval_steps_per_second": 0.279,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
"step": 8
|
45 |
},
|
46 |
{
|
47 |
+
"epoch": 4.8,
|
48 |
+
"eval_accuracy": 0.9714285714285714,
|
49 |
+
"eval_loss": 0.1735035479068756,
|
50 |
+
"eval_runtime": 6.91,
|
51 |
+
"eval_samples_per_second": 5.065,
|
52 |
+
"eval_steps_per_second": 0.289,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
53 |
"step": 10
|
54 |
},
|
55 |
{
|
56 |
+
"epoch": 5.8,
|
57 |
+
"eval_accuracy": 0.9714285714285714,
|
58 |
+
"eval_loss": 0.1692284345626831,
|
59 |
+
"eval_runtime": 6.8605,
|
60 |
+
"eval_samples_per_second": 5.102,
|
61 |
+
"eval_steps_per_second": 0.292,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
62 |
"step": 12
|
63 |
},
|
64 |
{
|
65 |
+
"epoch": 6.8,
|
66 |
+
"eval_accuracy": 0.9714285714285714,
|
67 |
+
"eval_loss": 0.1626075655221939,
|
68 |
+
"eval_runtime": 6.8284,
|
69 |
+
"eval_samples_per_second": 5.126,
|
70 |
+
"eval_steps_per_second": 0.293,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
71 |
"step": 14
|
72 |
},
|
73 |
{
|
74 |
+
"epoch": 7.8,
|
75 |
+
"eval_accuracy": 0.9714285714285714,
|
76 |
+
"eval_loss": 0.16586509346961975,
|
77 |
+
"eval_runtime": 6.3075,
|
78 |
+
"eval_samples_per_second": 5.549,
|
79 |
+
"eval_steps_per_second": 0.317,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
80 |
"step": 16
|
81 |
},
|
82 |
{
|
83 |
+
"epoch": 8.8,
|
84 |
+
"eval_accuracy": 0.9714285714285714,
|
85 |
+
"eval_loss": 0.1622493714094162,
|
86 |
+
"eval_runtime": 6.8493,
|
87 |
+
"eval_samples_per_second": 5.11,
|
88 |
+
"eval_steps_per_second": 0.292,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
89 |
"step": 18
|
90 |
},
|
91 |
{
|
92 |
+
"epoch": 9.8,
|
93 |
+
"learning_rate": 4.166666666666667e-05,
|
94 |
+
"loss": 0.2046,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
95 |
"step": 20
|
96 |
},
|
97 |
{
|
98 |
+
"epoch": 9.8,
|
99 |
+
"eval_accuracy": 0.9714285714285714,
|
100 |
+
"eval_loss": 0.1598205715417862,
|
101 |
+
"eval_runtime": 6.8584,
|
102 |
+
"eval_samples_per_second": 5.103,
|
103 |
+
"eval_steps_per_second": 0.292,
|
104 |
"step": 20
|
105 |
},
|
106 |
{
|
107 |
+
"epoch": 10.8,
|
108 |
+
"eval_accuracy": 0.9714285714285714,
|
109 |
+
"eval_loss": 0.16681121289730072,
|
110 |
+
"eval_runtime": 7.2746,
|
111 |
+
"eval_samples_per_second": 4.811,
|
112 |
+
"eval_steps_per_second": 0.275,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
113 |
"step": 22
|
114 |
},
|
115 |
{
|
116 |
+
"epoch": 11.8,
|
117 |
+
"eval_accuracy": 0.9714285714285714,
|
118 |
+
"eval_loss": 0.1746995747089386,
|
119 |
+
"eval_runtime": 7.1258,
|
120 |
+
"eval_samples_per_second": 4.912,
|
121 |
+
"eval_steps_per_second": 0.281,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
122 |
"step": 24
|
123 |
},
|
124 |
{
|
125 |
+
"epoch": 12.8,
|
126 |
+
"eval_accuracy": 0.9714285714285714,
|
127 |
+
"eval_loss": 0.18037545680999756,
|
128 |
+
"eval_runtime": 6.9863,
|
129 |
+
"eval_samples_per_second": 5.01,
|
130 |
+
"eval_steps_per_second": 0.286,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
131 |
"step": 26
|
132 |
},
|
133 |
{
|
134 |
+
"epoch": 13.8,
|
135 |
+
"eval_accuracy": 0.9714285714285714,
|
136 |
+
"eval_loss": 0.18370455503463745,
|
137 |
+
"eval_runtime": 6.1139,
|
138 |
+
"eval_samples_per_second": 5.725,
|
139 |
+
"eval_steps_per_second": 0.327,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
140 |
"step": 28
|
141 |
},
|
142 |
{
|
143 |
+
"epoch": 14.8,
|
144 |
+
"eval_accuracy": 0.9714285714285714,
|
145 |
+
"eval_loss": 0.1837157905101776,
|
146 |
+
"eval_runtime": 6.7444,
|
147 |
+
"eval_samples_per_second": 5.189,
|
148 |
+
"eval_steps_per_second": 0.297,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
149 |
"step": 30
|
150 |
},
|
151 |
{
|
152 |
+
"epoch": 15.8,
|
153 |
+
"eval_accuracy": 0.9714285714285714,
|
154 |
+
"eval_loss": 0.18111634254455566,
|
155 |
+
"eval_runtime": 7.0847,
|
156 |
+
"eval_samples_per_second": 4.94,
|
157 |
+
"eval_steps_per_second": 0.282,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
158 |
"step": 32
|
159 |
},
|
160 |
{
|
161 |
+
"epoch": 16.8,
|
162 |
+
"eval_accuracy": 0.9714285714285714,
|
163 |
+
"eval_loss": 0.18009454011917114,
|
164 |
+
"eval_runtime": 6.8211,
|
165 |
+
"eval_samples_per_second": 5.131,
|
166 |
+
"eval_steps_per_second": 0.293,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
167 |
"step": 34
|
168 |
},
|
169 |
{
|
170 |
+
"epoch": 17.8,
|
171 |
+
"eval_accuracy": 0.9714285714285714,
|
172 |
+
"eval_loss": 0.18408751487731934,
|
173 |
+
"eval_runtime": 6.9341,
|
174 |
+
"eval_samples_per_second": 5.048,
|
175 |
+
"eval_steps_per_second": 0.288,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
176 |
"step": 36
|
177 |
},
|
178 |
{
|
179 |
+
"epoch": 18.8,
|
180 |
+
"eval_accuracy": 0.9714285714285714,
|
181 |
+
"eval_loss": 0.1899442970752716,
|
182 |
+
"eval_runtime": 7.0358,
|
183 |
+
"eval_samples_per_second": 4.975,
|
184 |
+
"eval_steps_per_second": 0.284,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
185 |
"step": 38
|
186 |
},
|
187 |
{
|
188 |
+
"epoch": 19.8,
|
189 |
+
"learning_rate": 2.777777777777778e-05,
|
190 |
+
"loss": 0.1657,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
191 |
"step": 40
|
192 |
},
|
193 |
{
|
194 |
+
"epoch": 19.8,
|
195 |
+
"eval_accuracy": 0.9714285714285714,
|
196 |
+
"eval_loss": 0.19598019123077393,
|
197 |
+
"eval_runtime": 7.9601,
|
198 |
+
"eval_samples_per_second": 4.397,
|
199 |
+
"eval_steps_per_second": 0.251,
|
200 |
"step": 40
|
201 |
},
|
202 |
{
|
203 |
+
"epoch": 20.8,
|
204 |
+
"eval_accuracy": 0.9714285714285714,
|
205 |
+
"eval_loss": 0.19925238192081451,
|
206 |
+
"eval_runtime": 7.4131,
|
207 |
+
"eval_samples_per_second": 4.721,
|
208 |
+
"eval_steps_per_second": 0.27,
|
209 |
+
"step": 42
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 21.8,
|
213 |
+
"eval_accuracy": 0.9714285714285714,
|
214 |
+
"eval_loss": 0.20172713696956635,
|
215 |
+
"eval_runtime": 6.7891,
|
216 |
+
"eval_samples_per_second": 5.155,
|
217 |
+
"eval_steps_per_second": 0.295,
|
218 |
+
"step": 44
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 22.8,
|
222 |
+
"eval_accuracy": 0.9714285714285714,
|
223 |
+
"eval_loss": 0.2004331350326538,
|
224 |
+
"eval_runtime": 6.2296,
|
225 |
+
"eval_samples_per_second": 5.618,
|
226 |
+
"eval_steps_per_second": 0.321,
|
227 |
+
"step": 46
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 23.8,
|
231 |
+
"eval_accuracy": 0.9714285714285714,
|
232 |
+
"eval_loss": 0.19216616451740265,
|
233 |
+
"eval_runtime": 7.0857,
|
234 |
+
"eval_samples_per_second": 4.94,
|
235 |
+
"eval_steps_per_second": 0.282,
|
236 |
+
"step": 48
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 24.8,
|
240 |
+
"eval_accuracy": 0.9714285714285714,
|
241 |
+
"eval_loss": 0.18559373915195465,
|
242 |
+
"eval_runtime": 7.3612,
|
243 |
+
"eval_samples_per_second": 4.755,
|
244 |
+
"eval_steps_per_second": 0.272,
|
245 |
+
"step": 50
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 25.8,
|
249 |
+
"eval_accuracy": 0.9714285714285714,
|
250 |
+
"eval_loss": 0.18343603610992432,
|
251 |
+
"eval_runtime": 6.8088,
|
252 |
+
"eval_samples_per_second": 5.14,
|
253 |
+
"eval_steps_per_second": 0.294,
|
254 |
+
"step": 52
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 26.8,
|
258 |
+
"eval_accuracy": 0.9714285714285714,
|
259 |
+
"eval_loss": 0.18461596965789795,
|
260 |
+
"eval_runtime": 6.6873,
|
261 |
+
"eval_samples_per_second": 5.234,
|
262 |
+
"eval_steps_per_second": 0.299,
|
263 |
+
"step": 54
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 27.8,
|
267 |
+
"eval_accuracy": 0.9714285714285714,
|
268 |
+
"eval_loss": 0.18977026641368866,
|
269 |
+
"eval_runtime": 6.826,
|
270 |
+
"eval_samples_per_second": 5.127,
|
271 |
+
"eval_steps_per_second": 0.293,
|
272 |
+
"step": 56
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 28.8,
|
276 |
+
"eval_accuracy": 0.9714285714285714,
|
277 |
+
"eval_loss": 0.1951347291469574,
|
278 |
+
"eval_runtime": 6.8913,
|
279 |
+
"eval_samples_per_second": 5.079,
|
280 |
+
"eval_steps_per_second": 0.29,
|
281 |
+
"step": 58
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 29.8,
|
285 |
+
"learning_rate": 1.388888888888889e-05,
|
286 |
+
"loss": 0.1308,
|
287 |
+
"step": 60
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 29.8,
|
291 |
+
"eval_accuracy": 0.9714285714285714,
|
292 |
+
"eval_loss": 0.20185869932174683,
|
293 |
+
"eval_runtime": 6.8601,
|
294 |
+
"eval_samples_per_second": 5.102,
|
295 |
+
"eval_steps_per_second": 0.292,
|
296 |
+
"step": 60
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 30.8,
|
300 |
+
"eval_accuracy": 0.9714285714285714,
|
301 |
+
"eval_loss": 0.2095019370317459,
|
302 |
+
"eval_runtime": 7.2055,
|
303 |
+
"eval_samples_per_second": 4.857,
|
304 |
+
"eval_steps_per_second": 0.278,
|
305 |
+
"step": 62
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 31.8,
|
309 |
+
"eval_accuracy": 0.9714285714285714,
|
310 |
+
"eval_loss": 0.21445579826831818,
|
311 |
+
"eval_runtime": 7.4924,
|
312 |
+
"eval_samples_per_second": 4.671,
|
313 |
+
"eval_steps_per_second": 0.267,
|
314 |
+
"step": 64
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 32.8,
|
318 |
+
"eval_accuracy": 0.9714285714285714,
|
319 |
+
"eval_loss": 0.21541449427604675,
|
320 |
+
"eval_runtime": 6.8812,
|
321 |
+
"eval_samples_per_second": 5.086,
|
322 |
+
"eval_steps_per_second": 0.291,
|
323 |
+
"step": 66
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 33.8,
|
327 |
+
"eval_accuracy": 0.9714285714285714,
|
328 |
+
"eval_loss": 0.21372175216674805,
|
329 |
+
"eval_runtime": 6.9147,
|
330 |
+
"eval_samples_per_second": 5.062,
|
331 |
+
"eval_steps_per_second": 0.289,
|
332 |
+
"step": 68
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 34.8,
|
336 |
+
"eval_accuracy": 0.9714285714285714,
|
337 |
+
"eval_loss": 0.2116171419620514,
|
338 |
+
"eval_runtime": 7.0628,
|
339 |
+
"eval_samples_per_second": 4.956,
|
340 |
+
"eval_steps_per_second": 0.283,
|
341 |
+
"step": 70
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 35.8,
|
345 |
+
"eval_accuracy": 0.9714285714285714,
|
346 |
+
"eval_loss": 0.20960116386413574,
|
347 |
+
"eval_runtime": 7.1202,
|
348 |
+
"eval_samples_per_second": 4.916,
|
349 |
+
"eval_steps_per_second": 0.281,
|
350 |
+
"step": 72
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 36.8,
|
354 |
+
"eval_accuracy": 0.9714285714285714,
|
355 |
+
"eval_loss": 0.20841823518276215,
|
356 |
+
"eval_runtime": 7.1749,
|
357 |
+
"eval_samples_per_second": 4.878,
|
358 |
+
"eval_steps_per_second": 0.279,
|
359 |
+
"step": 74
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 37.8,
|
363 |
+
"eval_accuracy": 0.9714285714285714,
|
364 |
+
"eval_loss": 0.20780029892921448,
|
365 |
+
"eval_runtime": 6.9083,
|
366 |
+
"eval_samples_per_second": 5.066,
|
367 |
+
"eval_steps_per_second": 0.29,
|
368 |
+
"step": 76
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 38.8,
|
372 |
+
"eval_accuracy": 0.9714285714285714,
|
373 |
+
"eval_loss": 0.2074960172176361,
|
374 |
+
"eval_runtime": 6.9495,
|
375 |
+
"eval_samples_per_second": 5.036,
|
376 |
+
"eval_steps_per_second": 0.288,
|
377 |
+
"step": 78
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 39.8,
|
381 |
+
"learning_rate": 0.0,
|
382 |
+
"loss": 0.1053,
|
383 |
+
"step": 80
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 39.8,
|
387 |
+
"eval_accuracy": 0.9714285714285714,
|
388 |
+
"eval_loss": 0.20739802718162537,
|
389 |
+
"eval_runtime": 6.9547,
|
390 |
+
"eval_samples_per_second": 5.033,
|
391 |
+
"eval_steps_per_second": 0.288,
|
392 |
+
"step": 80
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 39.8,
|
396 |
+
"step": 80,
|
397 |
+
"total_flos": 3.088453228308726e+17,
|
398 |
+
"train_loss": 0.15159874260425568,
|
399 |
+
"train_runtime": 6474.8036,
|
400 |
+
"train_samples_per_second": 1.927,
|
401 |
+
"train_steps_per_second": 0.012
|
402 |
}
|
403 |
],
|
404 |
+
"max_steps": 80,
|
405 |
"num_train_epochs": 40,
|
406 |
+
"total_flos": 3.088453228308726e+17,
|
407 |
"trial_name": null,
|
408 |
"trial_params": null
|
409 |
}
|