Training in progress, epoch 4, checkpoint
Browse files
last-checkpoint/README.md
CHANGED
|
@@ -971,6 +971,46 @@ You can finetune this model on your own dataset.
|
|
| 971 |
| 2.9266 | 16380 | 1.0227 | - | - | - |
|
| 972 |
| 2.9516 | 16520 | 0.8159 | - | - | - |
|
| 973 |
| 2.9766 | 16660 | 0.8426 | - | - | - |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 974 |
|
| 975 |
</details>
|
| 976 |
|
|
|
|
| 971 |
| 2.9266 | 16380 | 1.0227 | - | - | - |
|
| 972 |
| 2.9516 | 16520 | 0.8159 | - | - | - |
|
| 973 |
| 2.9766 | 16660 | 0.8426 | - | - | - |
|
| 974 |
+
| 3.0016 | 16800 | 0.7955 | 1.1895 | 0.9681 | 0.4912 |
|
| 975 |
+
| 3.0266 | 16940 | 0.856 | - | - | - |
|
| 976 |
+
| 3.0516 | 17080 | 1.0754 | - | - | - |
|
| 977 |
+
| 3.0766 | 17220 | 0.9151 | - | - | - |
|
| 978 |
+
| 3.1017 | 17360 | 1.0051 | - | - | - |
|
| 979 |
+
| 3.1267 | 17500 | 1.0075 | 1.0658 | 0.9583 | 0.4853 |
|
| 980 |
+
| 3.1517 | 17640 | 0.9909 | - | - | - |
|
| 981 |
+
| 3.1767 | 17780 | 1.029 | - | - | - |
|
| 982 |
+
| 3.2017 | 17920 | 1.0292 | - | - | - |
|
| 983 |
+
| 3.2267 | 18060 | 0.8334 | - | - | - |
|
| 984 |
+
| 3.2517 | 18200 | 1.0119 | 1.1172 | 0.9485 | 0.4674 |
|
| 985 |
+
| 3.2768 | 18340 | 0.8582 | - | - | - |
|
| 986 |
+
| 3.3018 | 18480 | 1.0397 | - | - | - |
|
| 987 |
+
| 3.3268 | 18620 | 1.1988 | - | - | - |
|
| 988 |
+
| 3.3518 | 18760 | 0.9432 | - | - | - |
|
| 989 |
+
| 3.3768 | 18900 | 1.0573 | 1.0815 | 0.9437 | 0.4679 |
|
| 990 |
+
| 3.4018 | 19040 | 0.9829 | - | - | - |
|
| 991 |
+
| 3.4268 | 19180 | 1.0573 | - | - | - |
|
| 992 |
+
| 3.4518 | 19320 | 0.9449 | - | - | - |
|
| 993 |
+
| 3.4769 | 19460 | 1.2005 | - | - | - |
|
| 994 |
+
| 3.5019 | 19600 | 0.9171 | 1.2315 | 0.9503 | 0.4799 |
|
| 995 |
+
| 3.5269 | 19740 | 0.9425 | - | - | - |
|
| 996 |
+
| 3.5519 | 19880 | 1.1213 | - | - | - |
|
| 997 |
+
| 3.5769 | 20020 | 1.1128 | - | - | - |
|
| 998 |
+
| 3.6019 | 20160 | 1.331 | - | - | - |
|
| 999 |
+
| 3.6269 | 20300 | 1.0495 | 1.1413 | 0.9468 | 0.4434 |
|
| 1000 |
+
| 3.6520 | 20440 | 0.9698 | - | - | - |
|
| 1001 |
+
| 3.6770 | 20580 | 0.9148 | - | - | - |
|
| 1002 |
+
| 3.7020 | 20720 | 0.9042 | - | - | - |
|
| 1003 |
+
| 3.7270 | 20860 | 0.8232 | - | - | - |
|
| 1004 |
+
| 3.7520 | 21000 | 1.0163 | 1.2883 | 0.9020 | 0.4574 |
|
| 1005 |
+
| 3.7770 | 21140 | 0.9735 | - | - | - |
|
| 1006 |
+
| 3.8020 | 21280 | 0.8371 | - | - | - |
|
| 1007 |
+
| 3.8271 | 21420 | 0.6344 | - | - | - |
|
| 1008 |
+
| 3.8521 | 21560 | 0.87 | - | - | - |
|
| 1009 |
+
| 3.8771 | 21700 | 0.7404 | 1.0644 | 0.8661 | 0.4353 |
|
| 1010 |
+
| 3.9021 | 21840 | 0.8486 | - | - | - |
|
| 1011 |
+
| 3.9271 | 21980 | 0.8895 | - | - | - |
|
| 1012 |
+
| 3.9521 | 22120 | 0.7476 | - | - | - |
|
| 1013 |
+
| 3.9771 | 22260 | 0.6761 | - | - | - |
|
| 1014 |
|
| 1015 |
</details>
|
| 1016 |
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1130520122
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4345ef434b7ce58a400595f316107c458077d152783f92e11230dd06b00da241
|
| 3 |
size 1130520122
|
last-checkpoint/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 565251810
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c61e8dbee68a8b9aa8455be0f0f627f0276bc6305fcf58258493d0000b0b9ad
|
| 3 |
size 565251810
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:45081b173c6755e63dd0ede1a9b1a21b4183931e9d940410e69655e93a0291cf
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6243024eb0239246d61ff58dd2248929ed644805efadd2d460a5d88e0b9fcfa2
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 700,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1392,6 +1392,478 @@
|
|
| 1392 |
"learning_rate": 1.0994658958057889e-05,
|
| 1393 |
"loss": 0.8426,
|
| 1394 |
"step": 16660
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1395 |
}
|
| 1396 |
],
|
| 1397 |
"logging_steps": 140,
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
"eval_steps": 700,
|
| 6 |
+
"global_step": 22388,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1392 |
"learning_rate": 1.0994658958057889e-05,
|
| 1393 |
"loss": 0.8426,
|
| 1394 |
"step": 16660
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 3.0016080042880113,
|
| 1398 |
+
"grad_norm": 0.6402806043624878,
|
| 1399 |
+
"learning_rate": 1.0014032480000764e-05,
|
| 1400 |
+
"loss": 0.7955,
|
| 1401 |
+
"step": 16800
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 3.0016080042880113,
|
| 1405 |
+
"eval_nli-pairs_loss": 0.9680945873260498,
|
| 1406 |
+
"eval_nli-pairs_runtime": 13.331,
|
| 1407 |
+
"eval_nli-pairs_samples_per_second": 112.519,
|
| 1408 |
+
"eval_nli-pairs_steps_per_second": 4.726,
|
| 1409 |
+
"step": 16800
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 3.0016080042880113,
|
| 1413 |
+
"eval_scitail-pairs-pos_loss": 0.49118393659591675,
|
| 1414 |
+
"eval_scitail-pairs-pos_runtime": 15.2998,
|
| 1415 |
+
"eval_scitail-pairs-pos_samples_per_second": 85.23,
|
| 1416 |
+
"eval_scitail-pairs-pos_steps_per_second": 3.595,
|
| 1417 |
+
"step": 16800
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 3.0016080042880113,
|
| 1421 |
+
"eval_qnli-contrastive_loss": 1.1894794702529907,
|
| 1422 |
+
"eval_qnli-contrastive_runtime": 4.8432,
|
| 1423 |
+
"eval_qnli-contrastive_samples_per_second": 309.715,
|
| 1424 |
+
"eval_qnli-contrastive_steps_per_second": 13.008,
|
| 1425 |
+
"step": 16800
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 3.0266214043237447,
|
| 1429 |
+
"grad_norm": 3.155766010284424,
|
| 1430 |
+
"learning_rate": 9.03327071669702e-06,
|
| 1431 |
+
"loss": 0.856,
|
| 1432 |
+
"step": 16940
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 3.051634804359478,
|
| 1436 |
+
"grad_norm": 11.008296966552734,
|
| 1437 |
+
"learning_rate": 8.061829059993542e-06,
|
| 1438 |
+
"loss": 1.0754,
|
| 1439 |
+
"step": 17080
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 3.076648204395212,
|
| 1443 |
+
"grad_norm": 4.382720947265625,
|
| 1444 |
+
"learning_rate": 7.109073047846788e-06,
|
| 1445 |
+
"loss": 0.9151,
|
| 1446 |
+
"step": 17220
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 3.1016616044309453,
|
| 1450 |
+
"grad_norm": 2.755722761154175,
|
| 1451 |
+
"learning_rate": 6.184188072434878e-06,
|
| 1452 |
+
"loss": 1.0051,
|
| 1453 |
+
"step": 17360
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 3.1266750044666787,
|
| 1457 |
+
"grad_norm": 2.4547111988067627,
|
| 1458 |
+
"learning_rate": 5.296090825030854e-06,
|
| 1459 |
+
"loss": 1.0075,
|
| 1460 |
+
"step": 17500
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 3.1266750044666787,
|
| 1464 |
+
"eval_nli-pairs_loss": 0.9583492875099182,
|
| 1465 |
+
"eval_nli-pairs_runtime": 12.1773,
|
| 1466 |
+
"eval_nli-pairs_samples_per_second": 123.18,
|
| 1467 |
+
"eval_nli-pairs_steps_per_second": 5.174,
|
| 1468 |
+
"step": 17500
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 3.1266750044666787,
|
| 1472 |
+
"eval_scitail-pairs-pos_loss": 0.485266774892807,
|
| 1473 |
+
"eval_scitail-pairs-pos_runtime": 14.9222,
|
| 1474 |
+
"eval_scitail-pairs-pos_samples_per_second": 87.387,
|
| 1475 |
+
"eval_scitail-pairs-pos_steps_per_second": 3.686,
|
| 1476 |
+
"step": 17500
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 3.1266750044666787,
|
| 1480 |
+
"eval_qnli-contrastive_loss": 1.0658234357833862,
|
| 1481 |
+
"eval_qnli-contrastive_runtime": 4.7681,
|
| 1482 |
+
"eval_qnli-contrastive_samples_per_second": 314.592,
|
| 1483 |
+
"eval_qnli-contrastive_steps_per_second": 13.213,
|
| 1484 |
+
"step": 17500
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 3.151688404502412,
|
| 1488 |
+
"grad_norm": 19.061325073242188,
|
| 1489 |
+
"learning_rate": 4.453343331385006e-06,
|
| 1490 |
+
"loss": 0.9909,
|
| 1491 |
+
"step": 17640
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 3.1767018045381454,
|
| 1495 |
+
"grad_norm": 17.016021728515625,
|
| 1496 |
+
"learning_rate": 3.6640704063896858e-06,
|
| 1497 |
+
"loss": 1.029,
|
| 1498 |
+
"step": 17780
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 3.201715204573879,
|
| 1502 |
+
"grad_norm": 4.147863864898682,
|
| 1503 |
+
"learning_rate": 2.9358813238350816e-06,
|
| 1504 |
+
"loss": 1.0292,
|
| 1505 |
+
"step": 17920
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 3.226728604609612,
|
| 1509 |
+
"grad_norm": 27.60422706604004,
|
| 1510 |
+
"learning_rate": 2.275796456427173e-06,
|
| 1511 |
+
"loss": 0.8334,
|
| 1512 |
+
"step": 18060
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 3.2517420046453456,
|
| 1516 |
+
"grad_norm": 0.7800289392471313,
|
| 1517 |
+
"learning_rate": 1.6901795933215137e-06,
|
| 1518 |
+
"loss": 1.0119,
|
| 1519 |
+
"step": 18200
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 3.2517420046453456,
|
| 1523 |
+
"eval_nli-pairs_loss": 0.9484548568725586,
|
| 1524 |
+
"eval_nli-pairs_runtime": 12.0697,
|
| 1525 |
+
"eval_nli-pairs_samples_per_second": 124.279,
|
| 1526 |
+
"eval_nli-pairs_steps_per_second": 5.22,
|
| 1527 |
+
"step": 18200
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 3.2517420046453456,
|
| 1531 |
+
"eval_scitail-pairs-pos_loss": 0.4673975706100464,
|
| 1532 |
+
"eval_scitail-pairs-pos_runtime": 15.0509,
|
| 1533 |
+
"eval_scitail-pairs-pos_samples_per_second": 86.639,
|
| 1534 |
+
"eval_scitail-pairs-pos_steps_per_second": 3.654,
|
| 1535 |
+
"step": 18200
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 3.2517420046453456,
|
| 1539 |
+
"eval_qnli-contrastive_loss": 1.1171668767929077,
|
| 1540 |
+
"eval_qnli-contrastive_runtime": 4.7871,
|
| 1541 |
+
"eval_qnli-contrastive_samples_per_second": 313.345,
|
| 1542 |
+
"eval_qnli-contrastive_steps_per_second": 13.16,
|
| 1543 |
+
"step": 18200
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 3.2767554046810794,
|
| 1547 |
+
"grad_norm": 16.64696502685547,
|
| 1548 |
+
"learning_rate": 1.1846765876905709e-06,
|
| 1549 |
+
"loss": 0.8582,
|
| 1550 |
+
"step": 18340
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 3.3017688047168123,
|
| 1554 |
+
"grad_norm": 16.13783073425293,
|
| 1555 |
+
"learning_rate": 7.668532006209551e-07,
|
| 1556 |
+
"loss": 1.0397,
|
| 1557 |
+
"step": 18480
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 3.326782204752546,
|
| 1561 |
+
"grad_norm": 3.76619553565979,
|
| 1562 |
+
"learning_rate": 4.347306328421508e-07,
|
| 1563 |
+
"loss": 1.1988,
|
| 1564 |
+
"step": 18620
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 3.3517956047882795,
|
| 1568 |
+
"grad_norm": 10.401665687561035,
|
| 1569 |
+
"learning_rate": 1.948255365952012e-07,
|
| 1570 |
+
"loss": 0.9432,
|
| 1571 |
+
"step": 18760
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 3.376809004824013,
|
| 1575 |
+
"grad_norm": 2.400106191635132,
|
| 1576 |
+
"learning_rate": 4.945080454776929e-08,
|
| 1577 |
+
"loss": 1.0573,
|
| 1578 |
+
"step": 18900
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 3.376809004824013,
|
| 1582 |
+
"eval_nli-pairs_loss": 0.9437180757522583,
|
| 1583 |
+
"eval_nli-pairs_runtime": 12.0974,
|
| 1584 |
+
"eval_nli-pairs_samples_per_second": 123.993,
|
| 1585 |
+
"eval_nli-pairs_steps_per_second": 5.208,
|
| 1586 |
+
"step": 18900
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 3.376809004824013,
|
| 1590 |
+
"eval_scitail-pairs-pos_loss": 0.46788787841796875,
|
| 1591 |
+
"eval_scitail-pairs-pos_runtime": 15.1516,
|
| 1592 |
+
"eval_scitail-pairs-pos_samples_per_second": 86.063,
|
| 1593 |
+
"eval_scitail-pairs-pos_steps_per_second": 3.63,
|
| 1594 |
+
"step": 18900
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 3.376809004824013,
|
| 1598 |
+
"eval_qnli-contrastive_loss": 1.081482172012329,
|
| 1599 |
+
"eval_qnli-contrastive_runtime": 4.8096,
|
| 1600 |
+
"eval_qnli-contrastive_samples_per_second": 311.875,
|
| 1601 |
+
"eval_qnli-contrastive_steps_per_second": 13.099,
|
| 1602 |
+
"step": 18900
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 3.4018224048597463,
|
| 1606 |
+
"grad_norm": 25.33026695251465,
|
| 1607 |
+
"learning_rate": 7.974879220329356e-12,
|
| 1608 |
+
"loss": 0.9829,
|
| 1609 |
+
"step": 19040
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 3.4268358048954797,
|
| 1613 |
+
"grad_norm": 4.218173027038574,
|
| 1614 |
+
"learning_rate": 1.995302628075987e-05,
|
| 1615 |
+
"loss": 1.0573,
|
| 1616 |
+
"step": 19180
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 3.451849204931213,
|
| 1620 |
+
"grad_norm": 13.573431015014648,
|
| 1621 |
+
"learning_rate": 1.98101047527748e-05,
|
| 1622 |
+
"loss": 0.9449,
|
| 1623 |
+
"step": 19320
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 3.4768626049669464,
|
| 1627 |
+
"grad_norm": 6.658699989318848,
|
| 1628 |
+
"learning_rate": 1.9572605328335534e-05,
|
| 1629 |
+
"loss": 1.2005,
|
| 1630 |
+
"step": 19460
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 3.5018760050026803,
|
| 1634 |
+
"grad_norm": 6.075576305389404,
|
| 1635 |
+
"learning_rate": 1.924281770735239e-05,
|
| 1636 |
+
"loss": 0.9171,
|
| 1637 |
+
"step": 19600
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 3.5018760050026803,
|
| 1641 |
+
"eval_nli-pairs_loss": 0.9502684473991394,
|
| 1642 |
+
"eval_nli-pairs_runtime": 12.0413,
|
| 1643 |
+
"eval_nli-pairs_samples_per_second": 124.572,
|
| 1644 |
+
"eval_nli-pairs_steps_per_second": 5.232,
|
| 1645 |
+
"step": 19600
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 3.5018760050026803,
|
| 1649 |
+
"eval_scitail-pairs-pos_loss": 0.4798508584499359,
|
| 1650 |
+
"eval_scitail-pairs-pos_runtime": 14.9533,
|
| 1651 |
+
"eval_scitail-pairs-pos_samples_per_second": 87.205,
|
| 1652 |
+
"eval_scitail-pairs-pos_steps_per_second": 3.678,
|
| 1653 |
+
"step": 19600
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 3.5018760050026803,
|
| 1657 |
+
"eval_qnli-contrastive_loss": 1.2315282821655273,
|
| 1658 |
+
"eval_qnli-contrastive_runtime": 4.7188,
|
| 1659 |
+
"eval_qnli-contrastive_samples_per_second": 317.874,
|
| 1660 |
+
"eval_qnli-contrastive_steps_per_second": 13.351,
|
| 1661 |
+
"step": 19600
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 3.526889405038413,
|
| 1665 |
+
"grad_norm": 8.40775203704834,
|
| 1666 |
+
"learning_rate": 1.8823921327788075e-05,
|
| 1667 |
+
"loss": 0.9425,
|
| 1668 |
+
"step": 19740
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 3.551902805074147,
|
| 1672 |
+
"grad_norm": 11.214140892028809,
|
| 1673 |
+
"learning_rate": 1.831995471312526e-05,
|
| 1674 |
+
"loss": 1.1213,
|
| 1675 |
+
"step": 19880
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 3.5769162051098804,
|
| 1679 |
+
"grad_norm": 10.211651802062988,
|
| 1680 |
+
"learning_rate": 1.7735776537506483e-05,
|
| 1681 |
+
"loss": 1.1128,
|
| 1682 |
+
"step": 20020
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 3.6019296051456138,
|
| 1686 |
+
"grad_norm": 44.01512908935547,
|
| 1687 |
+
"learning_rate": 1.707701878391224e-05,
|
| 1688 |
+
"loss": 1.331,
|
| 1689 |
+
"step": 20160
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 3.626943005181347,
|
| 1693 |
+
"grad_norm": 13.295893669128418,
|
| 1694 |
+
"learning_rate": 1.6350032446972868e-05,
|
| 1695 |
+
"loss": 1.0495,
|
| 1696 |
+
"step": 20300
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 3.626943005181347,
|
| 1700 |
+
"eval_nli-pairs_loss": 0.9468088150024414,
|
| 1701 |
+
"eval_nli-pairs_runtime": 11.9325,
|
| 1702 |
+
"eval_nli-pairs_samples_per_second": 125.707,
|
| 1703 |
+
"eval_nli-pairs_steps_per_second": 5.28,
|
| 1704 |
+
"step": 20300
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 3.626943005181347,
|
| 1708 |
+
"eval_scitail-pairs-pos_loss": 0.4434490203857422,
|
| 1709 |
+
"eval_scitail-pairs-pos_runtime": 15.5134,
|
| 1710 |
+
"eval_scitail-pairs-pos_samples_per_second": 84.056,
|
| 1711 |
+
"eval_scitail-pairs-pos_steps_per_second": 3.545,
|
| 1712 |
+
"step": 20300
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 3.626943005181347,
|
| 1716 |
+
"eval_qnli-contrastive_loss": 1.141271710395813,
|
| 1717 |
+
"eval_qnli-contrastive_runtime": 4.7207,
|
| 1718 |
+
"eval_qnli-contrastive_samples_per_second": 317.752,
|
| 1719 |
+
"eval_qnli-contrastive_steps_per_second": 13.346,
|
| 1720 |
+
"step": 20300
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 3.6519564052170805,
|
| 1724 |
+
"grad_norm": 71.68439483642578,
|
| 1725 |
+
"learning_rate": 1.5561826303886085e-05,
|
| 1726 |
+
"loss": 0.9698,
|
| 1727 |
+
"step": 20440
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 3.676969805252814,
|
| 1731 |
+
"grad_norm": 5.957241058349609,
|
| 1732 |
+
"learning_rate": 1.4719999343741618e-05,
|
| 1733 |
+
"loss": 0.9148,
|
| 1734 |
+
"step": 20580
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 3.7019832052885473,
|
| 1738 |
+
"grad_norm": 1.4626597166061401,
|
| 1739 |
+
"learning_rate": 1.3839147028686583e-05,
|
| 1740 |
+
"loss": 0.9042,
|
| 1741 |
+
"step": 20720
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 3.726996605324281,
|
| 1745 |
+
"grad_norm": 2.4634809494018555,
|
| 1746 |
+
"learning_rate": 1.2915097668067934e-05,
|
| 1747 |
+
"loss": 0.8232,
|
| 1748 |
+
"step": 20860
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 3.752010005360014,
|
| 1752 |
+
"grad_norm": 1.5838899612426758,
|
| 1753 |
+
"learning_rate": 1.196294424410312e-05,
|
| 1754 |
+
"loss": 1.0163,
|
| 1755 |
+
"step": 21000
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 3.752010005360014,
|
| 1759 |
+
"eval_nli-pairs_loss": 0.9020450115203857,
|
| 1760 |
+
"eval_nli-pairs_runtime": 12.2572,
|
| 1761 |
+
"eval_nli-pairs_samples_per_second": 122.377,
|
| 1762 |
+
"eval_nli-pairs_steps_per_second": 5.14,
|
| 1763 |
+
"step": 21000
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 3.752010005360014,
|
| 1767 |
+
"eval_scitail-pairs-pos_loss": 0.4573577046394348,
|
| 1768 |
+
"eval_scitail-pairs-pos_runtime": 15.1478,
|
| 1769 |
+
"eval_scitail-pairs-pos_samples_per_second": 86.085,
|
| 1770 |
+
"eval_scitail-pairs-pos_steps_per_second": 3.631,
|
| 1771 |
+
"step": 21000
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 3.752010005360014,
|
| 1775 |
+
"eval_qnli-contrastive_loss": 1.2882591485977173,
|
| 1776 |
+
"eval_qnli-contrastive_runtime": 4.762,
|
| 1777 |
+
"eval_qnli-contrastive_samples_per_second": 314.992,
|
| 1778 |
+
"eval_qnli-contrastive_steps_per_second": 13.23,
|
| 1779 |
+
"step": 21000
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 3.777023405395748,
|
| 1783 |
+
"grad_norm": 5.878975868225098,
|
| 1784 |
+
"learning_rate": 1.099186633949893e-05,
|
| 1785 |
+
"loss": 0.9735,
|
| 1786 |
+
"step": 21140
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 3.8020368054314813,
|
| 1790 |
+
"grad_norm": 10.22749137878418,
|
| 1791 |
+
"learning_rate": 1.0011225985326909e-05,
|
| 1792 |
+
"loss": 0.8371,
|
| 1793 |
+
"step": 21280
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 3.8270502054672146,
|
| 1797 |
+
"grad_norm": 8.895988464355469,
|
| 1798 |
+
"learning_rate": 9.030477402944833e-06,
|
| 1799 |
+
"loss": 0.6344,
|
| 1800 |
+
"step": 21420
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 3.852063605502948,
|
| 1804 |
+
"grad_norm": 1.564530372619629,
|
| 1805 |
+
"learning_rate": 8.059075857124063e-06,
|
| 1806 |
+
"loss": 0.87,
|
| 1807 |
+
"step": 21560
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 3.8770770055386814,
|
| 1811 |
+
"grad_norm": 3.3526771068573,
|
| 1812 |
+
"learning_rate": 7.106386499117424e-06,
|
| 1813 |
+
"loss": 0.7404,
|
| 1814 |
+
"step": 21700
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 3.8770770055386814,
|
| 1818 |
+
"eval_nli-pairs_loss": 0.8661152720451355,
|
| 1819 |
+
"eval_nli-pairs_runtime": 11.9159,
|
| 1820 |
+
"eval_nli-pairs_samples_per_second": 125.883,
|
| 1821 |
+
"eval_nli-pairs_steps_per_second": 5.287,
|
| 1822 |
+
"step": 21700
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 3.8770770055386814,
|
| 1826 |
+
"eval_scitail-pairs-pos_loss": 0.4352877140045166,
|
| 1827 |
+
"eval_scitail-pairs-pos_runtime": 14.9412,
|
| 1828 |
+
"eval_scitail-pairs-pos_samples_per_second": 87.275,
|
| 1829 |
+
"eval_scitail-pairs-pos_steps_per_second": 3.681,
|
| 1830 |
+
"step": 21700
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 3.8770770055386814,
|
| 1834 |
+
"eval_qnli-contrastive_loss": 1.0643585920333862,
|
| 1835 |
+
"eval_qnli-contrastive_runtime": 4.7458,
|
| 1836 |
+
"eval_qnli-contrastive_samples_per_second": 316.066,
|
| 1837 |
+
"eval_qnli-contrastive_steps_per_second": 13.275,
|
| 1838 |
+
"step": 21700
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 3.9020904055744148,
|
| 1842 |
+
"grad_norm": 6.517562389373779,
|
| 1843 |
+
"learning_rate": 6.181594078499504e-06,
|
| 1844 |
+
"loss": 0.8486,
|
| 1845 |
+
"step": 21840
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 3.927103805610148,
|
| 1849 |
+
"grad_norm": 4.482045650482178,
|
| 1850 |
+
"learning_rate": 5.293614394235034e-06,
|
| 1851 |
+
"loss": 0.8895,
|
| 1852 |
+
"step": 21980
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 3.952117205645882,
|
| 1856 |
+
"grad_norm": 5.165999889373779,
|
| 1857 |
+
"learning_rate": 4.451008338663955e-06,
|
| 1858 |
+
"loss": 0.7476,
|
| 1859 |
+
"step": 22120
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 3.977130605681615,
|
| 1863 |
+
"grad_norm": 7.821371078491211,
|
| 1864 |
+
"learning_rate": 3.6618993630932396e-06,
|
| 1865 |
+
"loss": 0.6761,
|
| 1866 |
+
"step": 22260
|
| 1867 |
}
|
| 1868 |
],
|
| 1869 |
"logging_steps": 140,
|