Commit
•
46c0759
1
Parent(s):
7ac2142
Training in progress, step 3900
Browse files- {checkpoint-3200 → checkpoint-3800}/config.json +0 -0
- {checkpoint-3200 → checkpoint-3800}/optimizer.pt +1 -1
- {checkpoint-3200 → checkpoint-3800}/preprocessor_config.json +0 -0
- {checkpoint-3300 → checkpoint-3800}/pytorch_model.bin +1 -1
- {checkpoint-3300 → checkpoint-3800}/rng_state.pth +2 -2
- {checkpoint-3300 → checkpoint-3800}/scaler.pt +1 -1
- {checkpoint-3200 → checkpoint-3800}/scheduler.pt +1 -1
- {checkpoint-3300 → checkpoint-3800}/trainer_state.json +198 -3
- {checkpoint-3200 → checkpoint-3800}/training_args.bin +0 -0
- {checkpoint-3300 → checkpoint-3900}/config.json +0 -0
- {checkpoint-3300 → checkpoint-3900}/optimizer.pt +1 -1
- {checkpoint-3300 → checkpoint-3900}/preprocessor_config.json +0 -0
- {checkpoint-3200 → checkpoint-3900}/pytorch_model.bin +1 -1
- {checkpoint-3200 → checkpoint-3900}/rng_state.pth +2 -2
- {checkpoint-3200 → checkpoint-3900}/scaler.pt +1 -1
- {checkpoint-3300 → checkpoint-3900}/scheduler.pt +1 -1
- {checkpoint-3200 → checkpoint-3900}/trainer_state.json +276 -3
- {checkpoint-3300 → checkpoint-3900}/training_args.bin +0 -0
{checkpoint-3200 → checkpoint-3800}/config.json
RENAMED
|
File without changes
|
{checkpoint-3200 → checkpoint-3800}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2490337809
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3916c502476103f3ffc7a1308222895c30676a37dc692fc61cb70066e716d9ac
|
| 3 |
size 2490337809
|
{checkpoint-3200 → checkpoint-3800}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-3300 → checkpoint-3800}/pytorch_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262063089
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0b145b6b59bd869ef8bc9342a6f9c151b7c459935f0fd7b3877c9afe90a49de
|
| 3 |
size 1262063089
|
{checkpoint-3300 → checkpoint-3800}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd0c1e59d01355dc7800f78e2a8c707b3bf501a6ea7d92076b4d2400614623cf
|
| 3 |
+
size 14567
|
{checkpoint-3300 → checkpoint-3800}/scaler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 559
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:29eb923c97bb88614ccb0255ae678634f872dd8aa03ae16319e241eb7a1e8c90
|
| 3 |
size 559
|
{checkpoint-3200 → checkpoint-3800}/scheduler.pt
RENAMED
|
@@ -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 sha256:108cf3add24d85591d96de728715165debaf66f2fb85e7a11bb55ca6e478dd61
|
| 3 |
size 623
|
{checkpoint-3300 → checkpoint-3800}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -1293,11 +1293,206 @@
|
|
| 1293 |
"eval_steps_per_second": 0.791,
|
| 1294 |
"eval_wer": 0.1408458699971615,
|
| 1295 |
"step": 3300
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1296 |
}
|
| 1297 |
],
|
| 1298 |
"max_steps": 4550,
|
| 1299 |
"num_train_epochs": 50,
|
| 1300 |
-
"total_flos": 5.
|
| 1301 |
"trial_name": null,
|
| 1302 |
"trial_params": null
|
| 1303 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 41.75409836065574,
|
| 5 |
+
"global_step": 3800,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 1293 |
"eval_steps_per_second": 0.791,
|
| 1294 |
"eval_wer": 0.1408458699971615,
|
| 1295 |
"step": 3300
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 36.48,
|
| 1299 |
+
"learning_rate": 0.00012078431372549021,
|
| 1300 |
+
"loss": 0.772,
|
| 1301 |
+
"step": 3320
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 36.7,
|
| 1305 |
+
"learning_rate": 0.00011882352941176471,
|
| 1306 |
+
"loss": 0.7818,
|
| 1307 |
+
"step": 3340
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 36.92,
|
| 1311 |
+
"learning_rate": 0.00011696078431372549,
|
| 1312 |
+
"loss": 0.8016,
|
| 1313 |
+
"step": 3360
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 37.14,
|
| 1317 |
+
"learning_rate": 0.000115,
|
| 1318 |
+
"loss": 0.8061,
|
| 1319 |
+
"step": 3380
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 37.36,
|
| 1323 |
+
"learning_rate": 0.0001130392156862745,
|
| 1324 |
+
"loss": 0.7703,
|
| 1325 |
+
"step": 3400
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 37.36,
|
| 1329 |
+
"eval_loss": 0.16011376678943634,
|
| 1330 |
+
"eval_runtime": 187.5367,
|
| 1331 |
+
"eval_samples_per_second": 25.824,
|
| 1332 |
+
"eval_steps_per_second": 0.811,
|
| 1333 |
+
"eval_wer": 0.13692875390292364,
|
| 1334 |
+
"step": 3400
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 37.58,
|
| 1338 |
+
"learning_rate": 0.00011107843137254903,
|
| 1339 |
+
"loss": 0.7713,
|
| 1340 |
+
"step": 3420
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 37.8,
|
| 1344 |
+
"learning_rate": 0.00010911764705882353,
|
| 1345 |
+
"loss": 0.7712,
|
| 1346 |
+
"step": 3440
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 38.02,
|
| 1350 |
+
"learning_rate": 0.00010715686274509805,
|
| 1351 |
+
"loss": 0.8149,
|
| 1352 |
+
"step": 3460
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 38.24,
|
| 1356 |
+
"learning_rate": 0.00010519607843137255,
|
| 1357 |
+
"loss": 0.7885,
|
| 1358 |
+
"step": 3480
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 38.46,
|
| 1362 |
+
"learning_rate": 0.00010323529411764706,
|
| 1363 |
+
"loss": 0.7474,
|
| 1364 |
+
"step": 3500
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 38.46,
|
| 1368 |
+
"eval_loss": 0.1514146625995636,
|
| 1369 |
+
"eval_runtime": 190.8228,
|
| 1370 |
+
"eval_samples_per_second": 25.38,
|
| 1371 |
+
"eval_steps_per_second": 0.797,
|
| 1372 |
+
"eval_wer": 0.1342321884757309,
|
| 1373 |
+
"step": 3500
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 38.68,
|
| 1377 |
+
"learning_rate": 0.00010127450980392156,
|
| 1378 |
+
"loss": 0.785,
|
| 1379 |
+
"step": 3520
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 38.9,
|
| 1383 |
+
"learning_rate": 9.931372549019609e-05,
|
| 1384 |
+
"loss": 0.7677,
|
| 1385 |
+
"step": 3540
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 39.12,
|
| 1389 |
+
"learning_rate": 9.73529411764706e-05,
|
| 1390 |
+
"loss": 0.7849,
|
| 1391 |
+
"step": 3560
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 39.34,
|
| 1395 |
+
"learning_rate": 9.539215686274511e-05,
|
| 1396 |
+
"loss": 0.7637,
|
| 1397 |
+
"step": 3580
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 39.56,
|
| 1401 |
+
"learning_rate": 9.343137254901961e-05,
|
| 1402 |
+
"loss": 0.7719,
|
| 1403 |
+
"step": 3600
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 39.56,
|
| 1407 |
+
"eval_loss": 0.15932896733283997,
|
| 1408 |
+
"eval_runtime": 189.6806,
|
| 1409 |
+
"eval_samples_per_second": 25.532,
|
| 1410 |
+
"eval_steps_per_second": 0.801,
|
| 1411 |
+
"eval_wer": 0.1352540448481408,
|
| 1412 |
+
"step": 3600
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 39.78,
|
| 1416 |
+
"learning_rate": 9.147058823529412e-05,
|
| 1417 |
+
"loss": 0.7591,
|
| 1418 |
+
"step": 3620
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 39.99,
|
| 1422 |
+
"learning_rate": 8.950980392156862e-05,
|
| 1423 |
+
"loss": 0.7706,
|
| 1424 |
+
"step": 3640
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 40.22,
|
| 1428 |
+
"learning_rate": 8.754901960784314e-05,
|
| 1429 |
+
"loss": 0.7805,
|
| 1430 |
+
"step": 3660
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 40.44,
|
| 1434 |
+
"learning_rate": 8.558823529411765e-05,
|
| 1435 |
+
"loss": 0.7753,
|
| 1436 |
+
"step": 3680
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 40.66,
|
| 1440 |
+
"learning_rate": 8.362745098039217e-05,
|
| 1441 |
+
"loss": 0.7638,
|
| 1442 |
+
"step": 3700
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 40.66,
|
| 1446 |
+
"eval_loss": 0.15362653136253357,
|
| 1447 |
+
"eval_runtime": 192.1118,
|
| 1448 |
+
"eval_samples_per_second": 25.209,
|
| 1449 |
+
"eval_steps_per_second": 0.791,
|
| 1450 |
+
"eval_wer": 0.13380641498722678,
|
| 1451 |
+
"step": 3700
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 40.87,
|
| 1455 |
+
"learning_rate": 8.166666666666667e-05,
|
| 1456 |
+
"loss": 0.7791,
|
| 1457 |
+
"step": 3720
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 41.1,
|
| 1461 |
+
"learning_rate": 7.970588235294118e-05,
|
| 1462 |
+
"loss": 0.7931,
|
| 1463 |
+
"step": 3740
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 41.32,
|
| 1467 |
+
"learning_rate": 7.774509803921568e-05,
|
| 1468 |
+
"loss": 0.741,
|
| 1469 |
+
"step": 3760
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 41.54,
|
| 1473 |
+
"learning_rate": 7.57843137254902e-05,
|
| 1474 |
+
"loss": 0.7434,
|
| 1475 |
+
"step": 3780
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 41.75,
|
| 1479 |
+
"learning_rate": 7.38235294117647e-05,
|
| 1480 |
+
"loss": 0.771,
|
| 1481 |
+
"step": 3800
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 41.75,
|
| 1485 |
+
"eval_loss": 0.1530592143535614,
|
| 1486 |
+
"eval_runtime": 186.414,
|
| 1487 |
+
"eval_samples_per_second": 25.98,
|
| 1488 |
+
"eval_steps_per_second": 0.815,
|
| 1489 |
+
"eval_wer": 0.13170593244393983,
|
| 1490 |
+
"step": 3800
|
| 1491 |
}
|
| 1492 |
],
|
| 1493 |
"max_steps": 4550,
|
| 1494 |
"num_train_epochs": 50,
|
| 1495 |
+
"total_flos": 5.876357873041098e+19,
|
| 1496 |
"trial_name": null,
|
| 1497 |
"trial_params": null
|
| 1498 |
}
|
{checkpoint-3200 → checkpoint-3800}/training_args.bin
RENAMED
|
File without changes
|
{checkpoint-3300 → checkpoint-3900}/config.json
RENAMED
|
File without changes
|
{checkpoint-3300 → checkpoint-3900}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2490337809
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:168e59323a9bf241005906b1870612b1b524768b8a27757437012f8cdf781fc3
|
| 3 |
size 2490337809
|
{checkpoint-3300 → checkpoint-3900}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-3200 → checkpoint-3900}/pytorch_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262063089
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b689bc4f48a4fe8515030b196fb5bd4de5819b77555cbfd8e1979aa0d04c367
|
| 3 |
size 1262063089
|
{checkpoint-3200 → checkpoint-3900}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:266a5b6d767d7cc2ccd3807fca6c4cfb051cc7d3796a6836ef75e2cf3c6b3218
|
| 3 |
+
size 14567
|
{checkpoint-3200 → checkpoint-3900}/scaler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 559
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26f1e0a22621c69063b9f7d6715acac2896bf4745746722f322c93153a0c85b5
|
| 3 |
size 559
|
{checkpoint-3300 → checkpoint-3900}/scheduler.pt
RENAMED
|
@@ -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:e1bc34f9b39344288eae3f0b593523acd1748174f4d473bee33a329e62da9e8a
|
| 3 |
size 623
|
{checkpoint-3200 → checkpoint-3900}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -1254,11 +1254,284 @@
|
|
| 1254 |
"eval_steps_per_second": 0.795,
|
| 1255 |
"eval_wer": 0.1379222253760999,
|
| 1256 |
"step": 3200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1257 |
}
|
| 1258 |
],
|
| 1259 |
"max_steps": 4550,
|
| 1260 |
"num_train_epochs": 50,
|
| 1261 |
-
"total_flos":
|
| 1262 |
"trial_name": null,
|
| 1263 |
"trial_params": null
|
| 1264 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 42.85245901639344,
|
| 5 |
+
"global_step": 3900,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 1254 |
"eval_steps_per_second": 0.795,
|
| 1255 |
"eval_wer": 0.1379222253760999,
|
| 1256 |
"step": 3200
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 35.38,
|
| 1260 |
+
"learning_rate": 0.00013058823529411764,
|
| 1261 |
+
"loss": 0.7963,
|
| 1262 |
+
"step": 3220
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 35.6,
|
| 1266 |
+
"learning_rate": 0.00012862745098039216,
|
| 1267 |
+
"loss": 0.7895,
|
| 1268 |
+
"step": 3240
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 35.82,
|
| 1272 |
+
"learning_rate": 0.0001266666666666667,
|
| 1273 |
+
"loss": 0.7964,
|
| 1274 |
+
"step": 3260
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 36.04,
|
| 1278 |
+
"learning_rate": 0.0001247058823529412,
|
| 1279 |
+
"loss": 0.7931,
|
| 1280 |
+
"step": 3280
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 36.26,
|
| 1284 |
+
"learning_rate": 0.0001227450980392157,
|
| 1285 |
+
"loss": 0.7835,
|
| 1286 |
+
"step": 3300
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 36.26,
|
| 1290 |
+
"eval_loss": 0.16020993888378143,
|
| 1291 |
+
"eval_runtime": 192.1405,
|
| 1292 |
+
"eval_samples_per_second": 25.206,
|
| 1293 |
+
"eval_steps_per_second": 0.791,
|
| 1294 |
+
"eval_wer": 0.1408458699971615,
|
| 1295 |
+
"step": 3300
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 36.48,
|
| 1299 |
+
"learning_rate": 0.00012078431372549021,
|
| 1300 |
+
"loss": 0.772,
|
| 1301 |
+
"step": 3320
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 36.7,
|
| 1305 |
+
"learning_rate": 0.00011882352941176471,
|
| 1306 |
+
"loss": 0.7818,
|
| 1307 |
+
"step": 3340
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 36.92,
|
| 1311 |
+
"learning_rate": 0.00011696078431372549,
|
| 1312 |
+
"loss": 0.8016,
|
| 1313 |
+
"step": 3360
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 37.14,
|
| 1317 |
+
"learning_rate": 0.000115,
|
| 1318 |
+
"loss": 0.8061,
|
| 1319 |
+
"step": 3380
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 37.36,
|
| 1323 |
+
"learning_rate": 0.0001130392156862745,
|
| 1324 |
+
"loss": 0.7703,
|
| 1325 |
+
"step": 3400
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 37.36,
|
| 1329 |
+
"eval_loss": 0.16011376678943634,
|
| 1330 |
+
"eval_runtime": 187.5367,
|
| 1331 |
+
"eval_samples_per_second": 25.824,
|
| 1332 |
+
"eval_steps_per_second": 0.811,
|
| 1333 |
+
"eval_wer": 0.13692875390292364,
|
| 1334 |
+
"step": 3400
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 37.58,
|
| 1338 |
+
"learning_rate": 0.00011107843137254903,
|
| 1339 |
+
"loss": 0.7713,
|
| 1340 |
+
"step": 3420
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 37.8,
|
| 1344 |
+
"learning_rate": 0.00010911764705882353,
|
| 1345 |
+
"loss": 0.7712,
|
| 1346 |
+
"step": 3440
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 38.02,
|
| 1350 |
+
"learning_rate": 0.00010715686274509805,
|
| 1351 |
+
"loss": 0.8149,
|
| 1352 |
+
"step": 3460
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 38.24,
|
| 1356 |
+
"learning_rate": 0.00010519607843137255,
|
| 1357 |
+
"loss": 0.7885,
|
| 1358 |
+
"step": 3480
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 38.46,
|
| 1362 |
+
"learning_rate": 0.00010323529411764706,
|
| 1363 |
+
"loss": 0.7474,
|
| 1364 |
+
"step": 3500
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 38.46,
|
| 1368 |
+
"eval_loss": 0.1514146625995636,
|
| 1369 |
+
"eval_runtime": 190.8228,
|
| 1370 |
+
"eval_samples_per_second": 25.38,
|
| 1371 |
+
"eval_steps_per_second": 0.797,
|
| 1372 |
+
"eval_wer": 0.1342321884757309,
|
| 1373 |
+
"step": 3500
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 38.68,
|
| 1377 |
+
"learning_rate": 0.00010127450980392156,
|
| 1378 |
+
"loss": 0.785,
|
| 1379 |
+
"step": 3520
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 38.9,
|
| 1383 |
+
"learning_rate": 9.931372549019609e-05,
|
| 1384 |
+
"loss": 0.7677,
|
| 1385 |
+
"step": 3540
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 39.12,
|
| 1389 |
+
"learning_rate": 9.73529411764706e-05,
|
| 1390 |
+
"loss": 0.7849,
|
| 1391 |
+
"step": 3560
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 39.34,
|
| 1395 |
+
"learning_rate": 9.539215686274511e-05,
|
| 1396 |
+
"loss": 0.7637,
|
| 1397 |
+
"step": 3580
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 39.56,
|
| 1401 |
+
"learning_rate": 9.343137254901961e-05,
|
| 1402 |
+
"loss": 0.7719,
|
| 1403 |
+
"step": 3600
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 39.56,
|
| 1407 |
+
"eval_loss": 0.15932896733283997,
|
| 1408 |
+
"eval_runtime": 189.6806,
|
| 1409 |
+
"eval_samples_per_second": 25.532,
|
| 1410 |
+
"eval_steps_per_second": 0.801,
|
| 1411 |
+
"eval_wer": 0.1352540448481408,
|
| 1412 |
+
"step": 3600
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 39.78,
|
| 1416 |
+
"learning_rate": 9.147058823529412e-05,
|
| 1417 |
+
"loss": 0.7591,
|
| 1418 |
+
"step": 3620
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 39.99,
|
| 1422 |
+
"learning_rate": 8.950980392156862e-05,
|
| 1423 |
+
"loss": 0.7706,
|
| 1424 |
+
"step": 3640
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 40.22,
|
| 1428 |
+
"learning_rate": 8.754901960784314e-05,
|
| 1429 |
+
"loss": 0.7805,
|
| 1430 |
+
"step": 3660
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 40.44,
|
| 1434 |
+
"learning_rate": 8.558823529411765e-05,
|
| 1435 |
+
"loss": 0.7753,
|
| 1436 |
+
"step": 3680
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 40.66,
|
| 1440 |
+
"learning_rate": 8.362745098039217e-05,
|
| 1441 |
+
"loss": 0.7638,
|
| 1442 |
+
"step": 3700
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 40.66,
|
| 1446 |
+
"eval_loss": 0.15362653136253357,
|
| 1447 |
+
"eval_runtime": 192.1118,
|
| 1448 |
+
"eval_samples_per_second": 25.209,
|
| 1449 |
+
"eval_steps_per_second": 0.791,
|
| 1450 |
+
"eval_wer": 0.13380641498722678,
|
| 1451 |
+
"step": 3700
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 40.87,
|
| 1455 |
+
"learning_rate": 8.166666666666667e-05,
|
| 1456 |
+
"loss": 0.7791,
|
| 1457 |
+
"step": 3720
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 41.1,
|
| 1461 |
+
"learning_rate": 7.970588235294118e-05,
|
| 1462 |
+
"loss": 0.7931,
|
| 1463 |
+
"step": 3740
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 41.32,
|
| 1467 |
+
"learning_rate": 7.774509803921568e-05,
|
| 1468 |
+
"loss": 0.741,
|
| 1469 |
+
"step": 3760
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 41.54,
|
| 1473 |
+
"learning_rate": 7.57843137254902e-05,
|
| 1474 |
+
"loss": 0.7434,
|
| 1475 |
+
"step": 3780
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 41.75,
|
| 1479 |
+
"learning_rate": 7.38235294117647e-05,
|
| 1480 |
+
"loss": 0.771,
|
| 1481 |
+
"step": 3800
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 41.75,
|
| 1485 |
+
"eval_loss": 0.1530592143535614,
|
| 1486 |
+
"eval_runtime": 186.414,
|
| 1487 |
+
"eval_samples_per_second": 25.98,
|
| 1488 |
+
"eval_steps_per_second": 0.815,
|
| 1489 |
+
"eval_wer": 0.13170593244393983,
|
| 1490 |
+
"step": 3800
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 41.97,
|
| 1494 |
+
"learning_rate": 7.186274509803923e-05,
|
| 1495 |
+
"loss": 0.7765,
|
| 1496 |
+
"step": 3820
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 42.2,
|
| 1500 |
+
"learning_rate": 6.990196078431373e-05,
|
| 1501 |
+
"loss": 0.7599,
|
| 1502 |
+
"step": 3840
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 42.42,
|
| 1506 |
+
"learning_rate": 6.794117647058824e-05,
|
| 1507 |
+
"loss": 0.7782,
|
| 1508 |
+
"step": 3860
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 42.63,
|
| 1512 |
+
"learning_rate": 6.598039215686274e-05,
|
| 1513 |
+
"loss": 0.7395,
|
| 1514 |
+
"step": 3880
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 42.85,
|
| 1518 |
+
"learning_rate": 6.401960784313726e-05,
|
| 1519 |
+
"loss": 0.7594,
|
| 1520 |
+
"step": 3900
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 42.85,
|
| 1524 |
+
"eval_loss": 0.14983513951301575,
|
| 1525 |
+
"eval_runtime": 190.4439,
|
| 1526 |
+
"eval_samples_per_second": 25.43,
|
| 1527 |
+
"eval_steps_per_second": 0.798,
|
| 1528 |
+
"eval_wer": 0.12883905762134545,
|
| 1529 |
+
"step": 3900
|
| 1530 |
}
|
| 1531 |
],
|
| 1532 |
"max_steps": 4550,
|
| 1533 |
"num_train_epochs": 50,
|
| 1534 |
+
"total_flos": 6.032084123274907e+19,
|
| 1535 |
"trial_name": null,
|
| 1536 |
"trial_params": null
|
| 1537 |
}
|
{checkpoint-3300 → checkpoint-3900}/training_args.bin
RENAMED
|
File without changes
|