Training in progress, step 13500, checkpoint
Browse files
checkpoint-13500/model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1711848436
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e528f97b69e6f457368417e447376070b467d9e7951855b6a6f0b3ac0c3e6e26
|
3 |
size 1711848436
|
checkpoint-13500/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3424043887
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e21929b799a128bcc164c93bcbc6d98d034d262e7b501268725c49185ed73e06
|
3 |
size 3424043887
|
checkpoint-13500/rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1f9df87d270d2e2562b4f7d05ec0730338815d75ab95fb0a308f455fdc3875de
|
3 |
size 14503
|
checkpoint-13500/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0b4f83a21493b3fbd50dff6eb94efb9ac52f1b0d188ea215a01a2b7bda7c1d55
|
3 |
size 623
|
checkpoint-13500/trainer_state.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l58-l/checkpoint-13500",
|
4 |
-
"epoch":
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 13500,
|
7 |
"is_hyper_param_search": false,
|
@@ -9,413 +9,413 @@
|
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
-
"epoch": 0.
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate": 4.
|
15 |
-
"loss": 0.
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
-
"epoch": 0.
|
20 |
-
"eval_loss": 1.
|
21 |
-
"eval_runtime":
|
22 |
-
"eval_samples_per_second": 16.
|
23 |
-
"eval_steps_per_second": 2.
|
24 |
"step": 500
|
25 |
},
|
26 |
{
|
27 |
-
"epoch": 0.
|
28 |
-
"grad_norm":
|
29 |
-
"learning_rate": 4.
|
30 |
-
"loss": 0.
|
31 |
"step": 1000
|
32 |
},
|
33 |
{
|
34 |
-
"epoch": 0.
|
35 |
-
"eval_loss": 1.
|
36 |
-
"eval_runtime":
|
37 |
-
"eval_samples_per_second": 16.
|
38 |
-
"eval_steps_per_second": 2.
|
39 |
"step": 1000
|
40 |
},
|
41 |
{
|
42 |
-
"epoch": 0.
|
43 |
-
"grad_norm":
|
44 |
-
"learning_rate": 4.
|
45 |
-
"loss": 0.
|
46 |
"step": 1500
|
47 |
},
|
48 |
{
|
49 |
-
"epoch": 0.
|
50 |
-
"eval_loss": 1.
|
51 |
-
"eval_runtime":
|
52 |
-
"eval_samples_per_second": 15.
|
53 |
-
"eval_steps_per_second": 1.
|
54 |
"step": 1500
|
55 |
},
|
56 |
{
|
57 |
-
"epoch": 0.
|
58 |
-
"grad_norm":
|
59 |
-
"learning_rate": 4.
|
60 |
-
"loss": 0.
|
61 |
"step": 2000
|
62 |
},
|
63 |
{
|
64 |
-
"epoch": 0.
|
65 |
-
"eval_loss": 1.
|
66 |
-
"eval_runtime":
|
67 |
-
"eval_samples_per_second": 15.
|
68 |
"eval_steps_per_second": 1.962,
|
69 |
"step": 2000
|
70 |
},
|
71 |
{
|
72 |
-
"epoch": 0.
|
73 |
-
"grad_norm":
|
74 |
-
"learning_rate": 4.
|
75 |
-
"loss": 0.
|
76 |
"step": 2500
|
77 |
},
|
78 |
{
|
79 |
-
"epoch": 0.
|
80 |
-
"eval_loss": 1.
|
81 |
-
"eval_runtime":
|
82 |
-
"eval_samples_per_second": 15.
|
83 |
-
"eval_steps_per_second": 1.
|
84 |
"step": 2500
|
85 |
},
|
86 |
{
|
87 |
-
"epoch": 0.
|
88 |
-
"grad_norm":
|
89 |
-
"learning_rate": 4.
|
90 |
-
"loss": 0.
|
91 |
"step": 3000
|
92 |
},
|
93 |
{
|
94 |
-
"epoch": 0.
|
95 |
-
"eval_loss": 1.
|
96 |
-
"eval_runtime":
|
97 |
-
"eval_samples_per_second": 15.
|
98 |
-
"eval_steps_per_second": 1.
|
99 |
"step": 3000
|
100 |
},
|
101 |
{
|
102 |
-
"epoch": 0.
|
103 |
-
"grad_norm":
|
104 |
-
"learning_rate":
|
105 |
-
"loss": 0.
|
106 |
"step": 3500
|
107 |
},
|
108 |
{
|
109 |
-
"epoch": 0.
|
110 |
-
"eval_loss": 1.
|
111 |
-
"eval_runtime":
|
112 |
-
"eval_samples_per_second": 15.
|
113 |
-
"eval_steps_per_second": 1.
|
114 |
"step": 3500
|
115 |
},
|
116 |
{
|
117 |
-
"epoch": 0.
|
118 |
-
"grad_norm":
|
119 |
-
"learning_rate":
|
120 |
-
"loss": 0.
|
121 |
"step": 4000
|
122 |
},
|
123 |
{
|
124 |
-
"epoch": 0.
|
125 |
-
"eval_loss": 1.
|
126 |
-
"eval_runtime":
|
127 |
-
"eval_samples_per_second": 15.
|
128 |
-
"eval_steps_per_second": 1.
|
129 |
"step": 4000
|
130 |
},
|
131 |
{
|
132 |
-
"epoch": 0.
|
133 |
-
"grad_norm":
|
134 |
-
"learning_rate":
|
135 |
-
"loss": 0.
|
136 |
"step": 4500
|
137 |
},
|
138 |
{
|
139 |
-
"epoch": 0.
|
140 |
-
"eval_loss": 1.
|
141 |
-
"eval_runtime":
|
142 |
-
"eval_samples_per_second": 15.
|
143 |
-
"eval_steps_per_second": 1.
|
144 |
"step": 4500
|
145 |
},
|
146 |
{
|
147 |
-
"epoch": 0.
|
148 |
-
"grad_norm":
|
149 |
-
"learning_rate":
|
150 |
-
"loss": 0.
|
151 |
"step": 5000
|
152 |
},
|
153 |
{
|
154 |
-
"epoch": 0.
|
155 |
-
"eval_loss": 1.
|
156 |
-
"eval_runtime":
|
157 |
-
"eval_samples_per_second": 15.
|
158 |
-
"eval_steps_per_second": 1.
|
159 |
"step": 5000
|
160 |
},
|
161 |
{
|
162 |
-
"epoch":
|
163 |
-
"grad_norm": 0.
|
164 |
-
"learning_rate":
|
165 |
-
"loss": 0.
|
166 |
"step": 5500
|
167 |
},
|
168 |
{
|
169 |
-
"epoch":
|
170 |
-
"eval_loss": 1.
|
171 |
-
"eval_runtime":
|
172 |
-
"eval_samples_per_second":
|
173 |
-
"eval_steps_per_second":
|
174 |
"step": 5500
|
175 |
},
|
176 |
{
|
177 |
-
"epoch":
|
178 |
-
"grad_norm":
|
179 |
-
"learning_rate": 3.
|
180 |
-
"loss": 0.
|
181 |
"step": 6000
|
182 |
},
|
183 |
{
|
184 |
-
"epoch":
|
185 |
-
"eval_loss": 1.
|
186 |
-
"eval_runtime":
|
187 |
-
"eval_samples_per_second": 15.
|
188 |
-
"eval_steps_per_second": 1.
|
189 |
"step": 6000
|
190 |
},
|
191 |
{
|
192 |
-
"epoch":
|
193 |
-
"grad_norm":
|
194 |
-
"learning_rate": 3.
|
195 |
-
"loss": 0.
|
196 |
"step": 6500
|
197 |
},
|
198 |
{
|
199 |
-
"epoch":
|
200 |
-
"eval_loss": 1.
|
201 |
-
"eval_runtime":
|
202 |
-
"eval_samples_per_second": 15.
|
203 |
-
"eval_steps_per_second": 1.
|
204 |
"step": 6500
|
205 |
},
|
206 |
{
|
207 |
-
"epoch":
|
208 |
-
"grad_norm": 0.
|
209 |
-
"learning_rate":
|
210 |
-
"loss": 0.
|
211 |
"step": 7000
|
212 |
},
|
213 |
{
|
214 |
-
"epoch":
|
215 |
-
"eval_loss": 1.
|
216 |
-
"eval_runtime":
|
217 |
-
"eval_samples_per_second": 15.
|
218 |
-
"eval_steps_per_second": 1.
|
219 |
"step": 7000
|
220 |
},
|
221 |
{
|
222 |
-
"epoch":
|
223 |
-
"grad_norm":
|
224 |
-
"learning_rate":
|
225 |
-
"loss": 0.
|
226 |
"step": 7500
|
227 |
},
|
228 |
{
|
229 |
-
"epoch":
|
230 |
-
"eval_loss":
|
231 |
-
"eval_runtime":
|
232 |
-
"eval_samples_per_second": 15.
|
233 |
-
"eval_steps_per_second": 1.
|
234 |
"step": 7500
|
235 |
},
|
236 |
{
|
237 |
-
"epoch":
|
238 |
-
"grad_norm":
|
239 |
-
"learning_rate":
|
240 |
-
"loss": 0.
|
241 |
"step": 8000
|
242 |
},
|
243 |
{
|
244 |
-
"epoch":
|
245 |
-
"eval_loss":
|
246 |
-
"eval_runtime":
|
247 |
-
"eval_samples_per_second":
|
248 |
-
"eval_steps_per_second":
|
249 |
"step": 8000
|
250 |
},
|
251 |
{
|
252 |
-
"epoch":
|
253 |
-
"grad_norm":
|
254 |
-
"learning_rate":
|
255 |
-
"loss": 0.
|
256 |
"step": 8500
|
257 |
},
|
258 |
{
|
259 |
-
"epoch":
|
260 |
-
"eval_loss":
|
261 |
-
"eval_runtime":
|
262 |
-
"eval_samples_per_second": 15.
|
263 |
-
"eval_steps_per_second": 1.
|
264 |
"step": 8500
|
265 |
},
|
266 |
{
|
267 |
-
"epoch":
|
268 |
-
"grad_norm":
|
269 |
-
"learning_rate":
|
270 |
-
"loss": 0.
|
271 |
"step": 9000
|
272 |
},
|
273 |
{
|
274 |
-
"epoch":
|
275 |
-
"eval_loss":
|
276 |
-
"eval_runtime":
|
277 |
-
"eval_samples_per_second":
|
278 |
-
"eval_steps_per_second":
|
279 |
"step": 9000
|
280 |
},
|
281 |
{
|
282 |
-
"epoch": 1.
|
283 |
-
"grad_norm":
|
284 |
-
"learning_rate":
|
285 |
-
"loss": 0.
|
286 |
"step": 9500
|
287 |
},
|
288 |
{
|
289 |
-
"epoch": 1.
|
290 |
-
"eval_loss": 0.
|
291 |
-
"eval_runtime":
|
292 |
-
"eval_samples_per_second": 15.
|
293 |
-
"eval_steps_per_second":
|
294 |
"step": 9500
|
295 |
},
|
296 |
{
|
297 |
-
"epoch": 1.
|
298 |
-
"grad_norm":
|
299 |
-
"learning_rate":
|
300 |
-
"loss": 0.
|
301 |
"step": 10000
|
302 |
},
|
303 |
{
|
304 |
-
"epoch": 1.
|
305 |
-
"eval_loss": 0.
|
306 |
-
"eval_runtime":
|
307 |
-
"eval_samples_per_second": 15.
|
308 |
-
"eval_steps_per_second": 1.
|
309 |
"step": 10000
|
310 |
},
|
311 |
{
|
312 |
-
"epoch": 1.
|
313 |
-
"grad_norm":
|
314 |
-
"learning_rate":
|
315 |
-
"loss": 0.
|
316 |
"step": 10500
|
317 |
},
|
318 |
{
|
319 |
-
"epoch": 1.
|
320 |
-
"eval_loss": 0.
|
321 |
-
"eval_runtime":
|
322 |
-
"eval_samples_per_second":
|
323 |
-
"eval_steps_per_second":
|
324 |
"step": 10500
|
325 |
},
|
326 |
{
|
327 |
-
"epoch":
|
328 |
-
"grad_norm":
|
329 |
-
"learning_rate":
|
330 |
-
"loss": 0.
|
331 |
"step": 11000
|
332 |
},
|
333 |
{
|
334 |
-
"epoch":
|
335 |
-
"eval_loss": 0.
|
336 |
-
"eval_runtime":
|
337 |
-
"eval_samples_per_second": 15.
|
338 |
-
"eval_steps_per_second":
|
339 |
"step": 11000
|
340 |
},
|
341 |
{
|
342 |
-
"epoch":
|
343 |
-
"grad_norm":
|
344 |
-
"learning_rate":
|
345 |
-
"loss": 0.
|
346 |
"step": 11500
|
347 |
},
|
348 |
{
|
349 |
-
"epoch":
|
350 |
-
"eval_loss": 0.
|
351 |
-
"eval_runtime":
|
352 |
-
"eval_samples_per_second": 15.
|
353 |
-
"eval_steps_per_second": 1.
|
354 |
"step": 11500
|
355 |
},
|
356 |
{
|
357 |
-
"epoch":
|
358 |
-
"grad_norm":
|
359 |
-
"learning_rate":
|
360 |
-
"loss": 0.
|
361 |
"step": 12000
|
362 |
},
|
363 |
{
|
364 |
-
"epoch":
|
365 |
-
"eval_loss": 0.
|
366 |
-
"eval_runtime":
|
367 |
-
"eval_samples_per_second":
|
368 |
-
"eval_steps_per_second":
|
369 |
"step": 12000
|
370 |
},
|
371 |
{
|
372 |
-
"epoch":
|
373 |
-
"grad_norm":
|
374 |
-
"learning_rate":
|
375 |
-
"loss": 0.
|
376 |
"step": 12500
|
377 |
},
|
378 |
{
|
379 |
-
"epoch":
|
380 |
-
"eval_loss": 0.
|
381 |
-
"eval_runtime":
|
382 |
-
"eval_samples_per_second": 15.
|
383 |
-
"eval_steps_per_second": 1.
|
384 |
"step": 12500
|
385 |
},
|
386 |
{
|
387 |
-
"epoch":
|
388 |
-
"grad_norm": 0.
|
389 |
-
"learning_rate":
|
390 |
-
"loss": 0.
|
391 |
"step": 13000
|
392 |
},
|
393 |
{
|
394 |
-
"epoch":
|
395 |
-
"eval_loss": 0.
|
396 |
-
"eval_runtime":
|
397 |
-
"eval_samples_per_second": 15.
|
398 |
-
"eval_steps_per_second": 1.
|
399 |
"step": 13000
|
400 |
},
|
401 |
{
|
402 |
-
"epoch":
|
403 |
-
"grad_norm":
|
404 |
-
"learning_rate":
|
405 |
-
"loss": 0.
|
406 |
"step": 13500
|
407 |
},
|
408 |
{
|
409 |
-
"epoch":
|
410 |
-
"eval_loss": 0.
|
411 |
-
"eval_runtime":
|
412 |
-
"eval_samples_per_second":
|
413 |
-
"eval_steps_per_second":
|
414 |
"step": 13500
|
415 |
}
|
416 |
],
|
417 |
"logging_steps": 500,
|
418 |
-
"max_steps":
|
419 |
"num_input_tokens_seen": 0,
|
420 |
"num_train_epochs": 3,
|
421 |
"save_steps": 500,
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.9353039860725403,
|
3 |
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l58-l/checkpoint-13500",
|
4 |
+
"epoch": 1.4398464163822526,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 13500,
|
7 |
"is_hyper_param_search": false,
|
|
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
+
"epoch": 0.05332764505119454,
|
13 |
+
"grad_norm": 217.40109252929688,
|
14 |
+
"learning_rate": 4.911120591581342e-08,
|
15 |
+
"loss": 0.4667,
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
+
"epoch": 0.05332764505119454,
|
20 |
+
"eval_loss": 1.4427576065063477,
|
21 |
+
"eval_runtime": 60.328,
|
22 |
+
"eval_samples_per_second": 16.361,
|
23 |
+
"eval_steps_per_second": 2.055,
|
24 |
"step": 500
|
25 |
},
|
26 |
{
|
27 |
+
"epoch": 0.10665529010238908,
|
28 |
+
"grad_norm": 70.38941955566406,
|
29 |
+
"learning_rate": 4.822241183162685e-08,
|
30 |
+
"loss": 0.4534,
|
31 |
"step": 1000
|
32 |
},
|
33 |
{
|
34 |
+
"epoch": 0.10665529010238908,
|
35 |
+
"eval_loss": 1.3823094367980957,
|
36 |
+
"eval_runtime": 61.4884,
|
37 |
+
"eval_samples_per_second": 16.052,
|
38 |
+
"eval_steps_per_second": 2.017,
|
39 |
"step": 1000
|
40 |
},
|
41 |
{
|
42 |
+
"epoch": 0.1599829351535836,
|
43 |
+
"grad_norm": 389.9967956542969,
|
44 |
+
"learning_rate": 4.733361774744027e-08,
|
45 |
+
"loss": 0.3753,
|
46 |
"step": 1500
|
47 |
},
|
48 |
{
|
49 |
+
"epoch": 0.1599829351535836,
|
50 |
+
"eval_loss": 1.3328827619552612,
|
51 |
+
"eval_runtime": 63.0378,
|
52 |
+
"eval_samples_per_second": 15.657,
|
53 |
+
"eval_steps_per_second": 1.967,
|
54 |
"step": 1500
|
55 |
},
|
56 |
{
|
57 |
+
"epoch": 0.21331058020477817,
|
58 |
+
"grad_norm": 13.427445411682129,
|
59 |
+
"learning_rate": 4.6444823663253694e-08,
|
60 |
+
"loss": 0.3361,
|
61 |
"step": 2000
|
62 |
},
|
63 |
{
|
64 |
+
"epoch": 0.21331058020477817,
|
65 |
+
"eval_loss": 1.2922736406326294,
|
66 |
+
"eval_runtime": 63.1958,
|
67 |
+
"eval_samples_per_second": 15.618,
|
68 |
"eval_steps_per_second": 1.962,
|
69 |
"step": 2000
|
70 |
},
|
71 |
{
|
72 |
+
"epoch": 0.2666382252559727,
|
73 |
+
"grad_norm": 320.7197265625,
|
74 |
+
"learning_rate": 4.555602957906712e-08,
|
75 |
+
"loss": 0.3597,
|
76 |
"step": 2500
|
77 |
},
|
78 |
{
|
79 |
+
"epoch": 0.2666382252559727,
|
80 |
+
"eval_loss": 1.257211685180664,
|
81 |
+
"eval_runtime": 63.3241,
|
82 |
+
"eval_samples_per_second": 15.586,
|
83 |
+
"eval_steps_per_second": 1.958,
|
84 |
"step": 2500
|
85 |
},
|
86 |
{
|
87 |
+
"epoch": 0.3199658703071672,
|
88 |
+
"grad_norm": 0.0019184721168130636,
|
89 |
+
"learning_rate": 4.4667235494880546e-08,
|
90 |
+
"loss": 0.3042,
|
91 |
"step": 3000
|
92 |
},
|
93 |
{
|
94 |
+
"epoch": 0.3199658703071672,
|
95 |
+
"eval_loss": 1.2240594625473022,
|
96 |
+
"eval_runtime": 63.5118,
|
97 |
+
"eval_samples_per_second": 15.54,
|
98 |
+
"eval_steps_per_second": 1.952,
|
99 |
"step": 3000
|
100 |
},
|
101 |
{
|
102 |
+
"epoch": 0.37329351535836175,
|
103 |
+
"grad_norm": 11.738512992858887,
|
104 |
+
"learning_rate": 4.3778441410693965e-08,
|
105 |
+
"loss": 0.3252,
|
106 |
"step": 3500
|
107 |
},
|
108 |
{
|
109 |
+
"epoch": 0.37329351535836175,
|
110 |
+
"eval_loss": 1.1988952159881592,
|
111 |
+
"eval_runtime": 63.5384,
|
112 |
+
"eval_samples_per_second": 15.534,
|
113 |
+
"eval_steps_per_second": 1.952,
|
114 |
"step": 3500
|
115 |
},
|
116 |
{
|
117 |
+
"epoch": 0.42662116040955633,
|
118 |
+
"grad_norm": 5.839956283569336,
|
119 |
+
"learning_rate": 4.28896473265074e-08,
|
120 |
+
"loss": 0.2439,
|
121 |
"step": 4000
|
122 |
},
|
123 |
{
|
124 |
+
"epoch": 0.42662116040955633,
|
125 |
+
"eval_loss": 1.1730674505233765,
|
126 |
+
"eval_runtime": 62.3497,
|
127 |
+
"eval_samples_per_second": 15.83,
|
128 |
+
"eval_steps_per_second": 1.989,
|
129 |
"step": 4000
|
130 |
},
|
131 |
{
|
132 |
+
"epoch": 0.47994880546075086,
|
133 |
+
"grad_norm": 154.76502990722656,
|
134 |
+
"learning_rate": 4.2000853242320816e-08,
|
135 |
+
"loss": 0.3004,
|
136 |
"step": 4500
|
137 |
},
|
138 |
{
|
139 |
+
"epoch": 0.47994880546075086,
|
140 |
+
"eval_loss": 1.1497682332992554,
|
141 |
+
"eval_runtime": 63.5491,
|
142 |
+
"eval_samples_per_second": 15.531,
|
143 |
+
"eval_steps_per_second": 1.951,
|
144 |
"step": 4500
|
145 |
},
|
146 |
{
|
147 |
+
"epoch": 0.5332764505119454,
|
148 |
+
"grad_norm": 0.13150489330291748,
|
149 |
+
"learning_rate": 4.111205915813424e-08,
|
150 |
+
"loss": 0.3205,
|
151 |
"step": 5000
|
152 |
},
|
153 |
{
|
154 |
+
"epoch": 0.5332764505119454,
|
155 |
+
"eval_loss": 1.1258739233016968,
|
156 |
+
"eval_runtime": 62.1884,
|
157 |
+
"eval_samples_per_second": 15.871,
|
158 |
+
"eval_steps_per_second": 1.994,
|
159 |
"step": 5000
|
160 |
},
|
161 |
{
|
162 |
+
"epoch": 0.58660409556314,
|
163 |
+
"grad_norm": 0.3719326853752136,
|
164 |
+
"learning_rate": 4.022326507394767e-08,
|
165 |
+
"loss": 0.3091,
|
166 |
"step": 5500
|
167 |
},
|
168 |
{
|
169 |
+
"epoch": 0.58660409556314,
|
170 |
+
"eval_loss": 1.1053005456924438,
|
171 |
+
"eval_runtime": 63.437,
|
172 |
+
"eval_samples_per_second": 15.559,
|
173 |
+
"eval_steps_per_second": 1.955,
|
174 |
"step": 5500
|
175 |
},
|
176 |
{
|
177 |
+
"epoch": 0.6399317406143344,
|
178 |
+
"grad_norm": 170.89846801757812,
|
179 |
+
"learning_rate": 3.9334470989761093e-08,
|
180 |
+
"loss": 0.2657,
|
181 |
"step": 6000
|
182 |
},
|
183 |
{
|
184 |
+
"epoch": 0.6399317406143344,
|
185 |
+
"eval_loss": 1.0857468843460083,
|
186 |
+
"eval_runtime": 62.4055,
|
187 |
+
"eval_samples_per_second": 15.816,
|
188 |
+
"eval_steps_per_second": 1.987,
|
189 |
"step": 6000
|
190 |
},
|
191 |
{
|
192 |
+
"epoch": 0.693259385665529,
|
193 |
+
"grad_norm": 323.0997619628906,
|
194 |
+
"learning_rate": 3.844567690557451e-08,
|
195 |
+
"loss": 0.2251,
|
196 |
"step": 6500
|
197 |
},
|
198 |
{
|
199 |
+
"epoch": 0.693259385665529,
|
200 |
+
"eval_loss": 1.0695208311080933,
|
201 |
+
"eval_runtime": 63.1085,
|
202 |
+
"eval_samples_per_second": 15.64,
|
203 |
+
"eval_steps_per_second": 1.965,
|
204 |
"step": 6500
|
205 |
},
|
206 |
{
|
207 |
+
"epoch": 0.7465870307167235,
|
208 |
+
"grad_norm": 0.13254186511039734,
|
209 |
+
"learning_rate": 3.7556882821387945e-08,
|
210 |
+
"loss": 0.248,
|
211 |
"step": 7000
|
212 |
},
|
213 |
{
|
214 |
+
"epoch": 0.7465870307167235,
|
215 |
+
"eval_loss": 1.0535719394683838,
|
216 |
+
"eval_runtime": 62.3504,
|
217 |
+
"eval_samples_per_second": 15.83,
|
218 |
+
"eval_steps_per_second": 1.989,
|
219 |
"step": 7000
|
220 |
},
|
221 |
{
|
222 |
+
"epoch": 0.7999146757679181,
|
223 |
+
"grad_norm": 63.14700698852539,
|
224 |
+
"learning_rate": 3.6668088737201364e-08,
|
225 |
+
"loss": 0.2745,
|
226 |
"step": 7500
|
227 |
},
|
228 |
{
|
229 |
+
"epoch": 0.7999146757679181,
|
230 |
+
"eval_loss": 1.0388950109481812,
|
231 |
+
"eval_runtime": 62.2311,
|
232 |
+
"eval_samples_per_second": 15.86,
|
233 |
+
"eval_steps_per_second": 1.993,
|
234 |
"step": 7500
|
235 |
},
|
236 |
{
|
237 |
+
"epoch": 0.8532423208191127,
|
238 |
+
"grad_norm": 623.6299438476562,
|
239 |
+
"learning_rate": 3.577929465301478e-08,
|
240 |
+
"loss": 0.2857,
|
241 |
"step": 8000
|
242 |
},
|
243 |
{
|
244 |
+
"epoch": 0.8532423208191127,
|
245 |
+
"eval_loss": 1.0260424613952637,
|
246 |
+
"eval_runtime": 62.2792,
|
247 |
+
"eval_samples_per_second": 15.848,
|
248 |
+
"eval_steps_per_second": 1.991,
|
249 |
"step": 8000
|
250 |
},
|
251 |
{
|
252 |
+
"epoch": 0.9065699658703071,
|
253 |
+
"grad_norm": 155.54579162597656,
|
254 |
+
"learning_rate": 3.4890500568828216e-08,
|
255 |
+
"loss": 0.2365,
|
256 |
"step": 8500
|
257 |
},
|
258 |
{
|
259 |
+
"epoch": 0.9065699658703071,
|
260 |
+
"eval_loss": 1.0136038064956665,
|
261 |
+
"eval_runtime": 62.583,
|
262 |
+
"eval_samples_per_second": 15.771,
|
263 |
+
"eval_steps_per_second": 1.981,
|
264 |
"step": 8500
|
265 |
},
|
266 |
{
|
267 |
+
"epoch": 0.9598976109215017,
|
268 |
+
"grad_norm": 76.14079284667969,
|
269 |
+
"learning_rate": 3.4001706484641635e-08,
|
270 |
+
"loss": 0.2382,
|
271 |
"step": 9000
|
272 |
},
|
273 |
{
|
274 |
+
"epoch": 0.9598976109215017,
|
275 |
+
"eval_loss": 1.0034172534942627,
|
276 |
+
"eval_runtime": 63.5592,
|
277 |
+
"eval_samples_per_second": 15.529,
|
278 |
+
"eval_steps_per_second": 1.951,
|
279 |
"step": 9000
|
280 |
},
|
281 |
{
|
282 |
+
"epoch": 1.0132252559726962,
|
283 |
+
"grad_norm": 571.6434936523438,
|
284 |
+
"learning_rate": 3.311291240045506e-08,
|
285 |
+
"loss": 0.2419,
|
286 |
"step": 9500
|
287 |
},
|
288 |
{
|
289 |
+
"epoch": 1.0132252559726962,
|
290 |
+
"eval_loss": 0.994463324546814,
|
291 |
+
"eval_runtime": 61.8407,
|
292 |
+
"eval_samples_per_second": 15.96,
|
293 |
+
"eval_steps_per_second": 2.005,
|
294 |
"step": 9500
|
295 |
},
|
296 |
{
|
297 |
+
"epoch": 1.0665529010238908,
|
298 |
+
"grad_norm": 540.1460571289062,
|
299 |
+
"learning_rate": 3.2224118316268486e-08,
|
300 |
+
"loss": 0.2047,
|
301 |
"step": 10000
|
302 |
},
|
303 |
{
|
304 |
+
"epoch": 1.0665529010238908,
|
305 |
+
"eval_loss": 0.9851205348968506,
|
306 |
+
"eval_runtime": 62.1686,
|
307 |
+
"eval_samples_per_second": 15.876,
|
308 |
+
"eval_steps_per_second": 1.995,
|
309 |
"step": 10000
|
310 |
},
|
311 |
{
|
312 |
+
"epoch": 1.1198805460750854,
|
313 |
+
"grad_norm": 6.2389655113220215,
|
314 |
+
"learning_rate": 3.133532423208191e-08,
|
315 |
+
"loss": 0.2133,
|
316 |
"step": 10500
|
317 |
},
|
318 |
{
|
319 |
+
"epoch": 1.1198805460750854,
|
320 |
+
"eval_loss": 0.9759647250175476,
|
321 |
+
"eval_runtime": 62.8691,
|
322 |
+
"eval_samples_per_second": 15.699,
|
323 |
+
"eval_steps_per_second": 1.972,
|
324 |
"step": 10500
|
325 |
},
|
326 |
{
|
327 |
+
"epoch": 1.17320819112628,
|
328 |
+
"grad_norm": 321.5069885253906,
|
329 |
+
"learning_rate": 3.044653014789533e-08,
|
330 |
+
"loss": 0.207,
|
331 |
"step": 11000
|
332 |
},
|
333 |
{
|
334 |
+
"epoch": 1.17320819112628,
|
335 |
+
"eval_loss": 0.9683472514152527,
|
336 |
+
"eval_runtime": 61.915,
|
337 |
+
"eval_samples_per_second": 15.941,
|
338 |
+
"eval_steps_per_second": 2.003,
|
339 |
"step": 11000
|
340 |
},
|
341 |
{
|
342 |
+
"epoch": 1.2265358361774745,
|
343 |
+
"grad_norm": 0.0028632767498493195,
|
344 |
+
"learning_rate": 2.955773606370876e-08,
|
345 |
+
"loss": 0.2369,
|
346 |
"step": 11500
|
347 |
},
|
348 |
{
|
349 |
+
"epoch": 1.2265358361774745,
|
350 |
+
"eval_loss": 0.9597958922386169,
|
351 |
+
"eval_runtime": 62.4216,
|
352 |
+
"eval_samples_per_second": 15.812,
|
353 |
+
"eval_steps_per_second": 1.986,
|
354 |
"step": 11500
|
355 |
},
|
356 |
{
|
357 |
+
"epoch": 1.2798634812286689,
|
358 |
+
"grad_norm": 1.046075701713562,
|
359 |
+
"learning_rate": 2.8668941979522183e-08,
|
360 |
+
"loss": 0.1834,
|
361 |
"step": 12000
|
362 |
},
|
363 |
{
|
364 |
+
"epoch": 1.2798634812286689,
|
365 |
+
"eval_loss": 0.9528058767318726,
|
366 |
+
"eval_runtime": 62.9687,
|
367 |
+
"eval_samples_per_second": 15.674,
|
368 |
+
"eval_steps_per_second": 1.969,
|
369 |
"step": 12000
|
370 |
},
|
371 |
{
|
372 |
+
"epoch": 1.3331911262798635,
|
373 |
+
"grad_norm": 0.01093329768627882,
|
374 |
+
"learning_rate": 2.778014789533561e-08,
|
375 |
+
"loss": 0.2002,
|
376 |
"step": 12500
|
377 |
},
|
378 |
{
|
379 |
+
"epoch": 1.3331911262798635,
|
380 |
+
"eval_loss": 0.9465218782424927,
|
381 |
+
"eval_runtime": 62.0222,
|
382 |
+
"eval_samples_per_second": 15.914,
|
383 |
+
"eval_steps_per_second": 1.999,
|
384 |
"step": 12500
|
385 |
},
|
386 |
{
|
387 |
+
"epoch": 1.386518771331058,
|
388 |
+
"grad_norm": 0.13472874462604523,
|
389 |
+
"learning_rate": 2.689135381114903e-08,
|
390 |
+
"loss": 0.1918,
|
391 |
"step": 13000
|
392 |
},
|
393 |
{
|
394 |
+
"epoch": 1.386518771331058,
|
395 |
+
"eval_loss": 0.9417626857757568,
|
396 |
+
"eval_runtime": 62.4603,
|
397 |
+
"eval_samples_per_second": 15.802,
|
398 |
+
"eval_steps_per_second": 1.985,
|
399 |
"step": 13000
|
400 |
},
|
401 |
{
|
402 |
+
"epoch": 1.4398464163822526,
|
403 |
+
"grad_norm": 19.54829216003418,
|
404 |
+
"learning_rate": 2.600255972696246e-08,
|
405 |
+
"loss": 0.2264,
|
406 |
"step": 13500
|
407 |
},
|
408 |
{
|
409 |
+
"epoch": 1.4398464163822526,
|
410 |
+
"eval_loss": 0.9353039860725403,
|
411 |
+
"eval_runtime": 63.2324,
|
412 |
+
"eval_samples_per_second": 15.609,
|
413 |
+
"eval_steps_per_second": 1.961,
|
414 |
"step": 13500
|
415 |
}
|
416 |
],
|
417 |
"logging_steps": 500,
|
418 |
+
"max_steps": 28128,
|
419 |
"num_input_tokens_seen": 0,
|
420 |
"num_train_epochs": 3,
|
421 |
"save_steps": 500,
|
checkpoint-13500/training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4847
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:590a3eed3f4b2ba9fcb0b831d5e4c3f2410a46be8f57155ebf9fdd97c6c96524
|
3 |
size 4847
|