Commit
·
dab7aaa
1
Parent(s):
c59800f
Training in progress, step 24000
Browse files- adapter_model.bin +1 -1
- {checkpoint-20000 → checkpoint-23000/adapter_model}/README.md +0 -0
- {checkpoint-20000 → checkpoint-23000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-20000 → checkpoint-23000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-20000/adapter_model → checkpoint-24000}/README.md +0 -0
- {checkpoint-20000/adapter_model → checkpoint-24000}/adapter_config.json +0 -0
- {checkpoint-20000/adapter_model → checkpoint-24000}/adapter_model.bin +1 -1
- {checkpoint-20000 → checkpoint-24000}/optimizer.pt +1 -1
- {checkpoint-20000 → checkpoint-24000}/rng_state.pth +1 -1
- {checkpoint-20000 → checkpoint-24000}/scheduler.pt +1 -1
- {checkpoint-20000 → checkpoint-24000}/trainer_state.json +277 -5
- {checkpoint-20000 → checkpoint-24000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e5f593687f6b7f35b8db5be08cd3d3eabc5b69e4a5f70590ddb55cde4f73c64
|
| 3 |
size 500897101
|
{checkpoint-20000 → checkpoint-23000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-20000 → checkpoint-23000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-20000 → checkpoint-23000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b09fbcc84f52583b6b795b011e1c5f0c43a3db60e66b9f20591fcc7e4f4c5a5c
|
| 3 |
size 500897101
|
{checkpoint-20000/adapter_model → checkpoint-24000}/README.md
RENAMED
|
File without changes
|
{checkpoint-20000/adapter_model → checkpoint-24000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-20000/adapter_model → checkpoint-24000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e5f593687f6b7f35b8db5be08cd3d3eabc5b69e4a5f70590ddb55cde4f73c64
|
| 3 |
size 500897101
|
{checkpoint-20000 → checkpoint-24000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41c516a1cda2f637391a4cc14345088f2c9231c0a65bc15aa25db213427b391e
|
| 3 |
size 1001723453
|
{checkpoint-20000 → checkpoint-24000}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6a0c20bcdf319a5ffb4a98f7ca699fd1efef2927f5c6f105ca13297f084ba02
|
| 3 |
size 14575
|
{checkpoint-20000 → checkpoint-24000}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd7a0be07ab26faaad77ee68b49800320fcb18c774f44c984bf819516ecfa521
|
| 3 |
size 627
|
{checkpoint-20000 → checkpoint-24000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch": 0.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -1366,11 +1366,283 @@
|
|
| 1366 |
"eval_samples_per_second": 0.428,
|
| 1367 |
"eval_steps_per_second": 0.428,
|
| 1368 |
"step": 20000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1369 |
}
|
| 1370 |
],
|
| 1371 |
"max_steps": 80463,
|
| 1372 |
"num_train_epochs": 3,
|
| 1373 |
-
"total_flos":
|
| 1374 |
"trial_name": null,
|
| 1375 |
"trial_params": null
|
| 1376 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.6122664213180542,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-24000",
|
| 4 |
+
"epoch": 0.8948212221766526,
|
| 5 |
+
"global_step": 24000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 1366 |
"eval_samples_per_second": 0.428,
|
| 1367 |
"eval_steps_per_second": 0.428,
|
| 1368 |
"step": 20000
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.75,
|
| 1372 |
+
"learning_rate": 0.0001707748460945171,
|
| 1373 |
+
"loss": 0.5953,
|
| 1374 |
+
"step": 20100
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.75,
|
| 1378 |
+
"learning_rate": 0.0001704984413287202,
|
| 1379 |
+
"loss": 0.6329,
|
| 1380 |
+
"step": 20200
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.76,
|
| 1384 |
+
"learning_rate": 0.00017022096159817493,
|
| 1385 |
+
"loss": 0.6227,
|
| 1386 |
+
"step": 20300
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 0.76,
|
| 1390 |
+
"learning_rate": 0.00016994241113391003,
|
| 1391 |
+
"loss": 0.6022,
|
| 1392 |
+
"step": 20400
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 0.76,
|
| 1396 |
+
"learning_rate": 0.0001696627941832808,
|
| 1397 |
+
"loss": 0.604,
|
| 1398 |
+
"step": 20500
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 0.77,
|
| 1402 |
+
"learning_rate": 0.0001693821150099044,
|
| 1403 |
+
"loss": 0.6101,
|
| 1404 |
+
"step": 20600
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.77,
|
| 1408 |
+
"learning_rate": 0.00016910037789359485,
|
| 1409 |
+
"loss": 0.6242,
|
| 1410 |
+
"step": 20700
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.78,
|
| 1414 |
+
"learning_rate": 0.00016881758713029776,
|
| 1415 |
+
"loss": 0.6096,
|
| 1416 |
+
"step": 20800
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.78,
|
| 1420 |
+
"learning_rate": 0.0001685337470320248,
|
| 1421 |
+
"loss": 0.5948,
|
| 1422 |
+
"step": 20900
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 0.78,
|
| 1426 |
+
"learning_rate": 0.0001682488619267879,
|
| 1427 |
+
"loss": 0.5911,
|
| 1428 |
+
"step": 21000
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 0.78,
|
| 1432 |
+
"eval_loss": 0.6282580494880676,
|
| 1433 |
+
"eval_runtime": 1313.1215,
|
| 1434 |
+
"eval_samples_per_second": 0.413,
|
| 1435 |
+
"eval_steps_per_second": 0.413,
|
| 1436 |
+
"step": 21000
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 0.79,
|
| 1440 |
+
"learning_rate": 0.0001679629361585335,
|
| 1441 |
+
"loss": 0.5716,
|
| 1442 |
+
"step": 21100
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 0.79,
|
| 1446 |
+
"learning_rate": 0.00016767597408707594,
|
| 1447 |
+
"loss": 0.5957,
|
| 1448 |
+
"step": 21200
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 0.79,
|
| 1452 |
+
"learning_rate": 0.00016738798008803128,
|
| 1453 |
+
"loss": 0.6308,
|
| 1454 |
+
"step": 21300
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 0.8,
|
| 1458 |
+
"learning_rate": 0.00016709895855275048,
|
| 1459 |
+
"loss": 0.5891,
|
| 1460 |
+
"step": 21400
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.8,
|
| 1464 |
+
"learning_rate": 0.00016680891388825243,
|
| 1465 |
+
"loss": 0.6104,
|
| 1466 |
+
"step": 21500
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.81,
|
| 1470 |
+
"learning_rate": 0.00016651785051715674,
|
| 1471 |
+
"loss": 0.6344,
|
| 1472 |
+
"step": 21600
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.81,
|
| 1476 |
+
"learning_rate": 0.0001662257728776163,
|
| 1477 |
+
"loss": 0.604,
|
| 1478 |
+
"step": 21700
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 0.81,
|
| 1482 |
+
"learning_rate": 0.0001659326854232497,
|
| 1483 |
+
"loss": 0.6066,
|
| 1484 |
+
"step": 21800
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 0.82,
|
| 1488 |
+
"learning_rate": 0.0001656385926230732,
|
| 1489 |
+
"loss": 0.6324,
|
| 1490 |
+
"step": 21900
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 0.82,
|
| 1494 |
+
"learning_rate": 0.00016534349896143264,
|
| 1495 |
+
"loss": 0.5819,
|
| 1496 |
+
"step": 22000
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 0.82,
|
| 1500 |
+
"eval_loss": 0.6218891143798828,
|
| 1501 |
+
"eval_runtime": 1296.6038,
|
| 1502 |
+
"eval_samples_per_second": 0.418,
|
| 1503 |
+
"eval_steps_per_second": 0.418,
|
| 1504 |
+
"step": 22000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 0.82,
|
| 1508 |
+
"learning_rate": 0.00016504740893793512,
|
| 1509 |
+
"loss": 0.6145,
|
| 1510 |
+
"step": 22100
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 0.83,
|
| 1514 |
+
"learning_rate": 0.00016475032706738023,
|
| 1515 |
+
"loss": 0.6109,
|
| 1516 |
+
"step": 22200
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 0.83,
|
| 1520 |
+
"learning_rate": 0.0001644522578796914,
|
| 1521 |
+
"loss": 0.608,
|
| 1522 |
+
"step": 22300
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.84,
|
| 1526 |
+
"learning_rate": 0.0001641532059198466,
|
| 1527 |
+
"loss": 0.565,
|
| 1528 |
+
"step": 22400
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.84,
|
| 1532 |
+
"learning_rate": 0.00016385317574780942,
|
| 1533 |
+
"loss": 0.6139,
|
| 1534 |
+
"step": 22500
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 0.84,
|
| 1538 |
+
"learning_rate": 0.000163552171938459,
|
| 1539 |
+
"loss": 0.5888,
|
| 1540 |
+
"step": 22600
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 0.85,
|
| 1544 |
+
"learning_rate": 0.00016325019908152078,
|
| 1545 |
+
"loss": 0.6065,
|
| 1546 |
+
"step": 22700
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 0.85,
|
| 1550 |
+
"learning_rate": 0.0001629472617814962,
|
| 1551 |
+
"loss": 0.5959,
|
| 1552 |
+
"step": 22800
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 0.85,
|
| 1556 |
+
"learning_rate": 0.00016264336465759258,
|
| 1557 |
+
"loss": 0.5918,
|
| 1558 |
+
"step": 22900
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 0.86,
|
| 1562 |
+
"learning_rate": 0.0001623385123436528,
|
| 1563 |
+
"loss": 0.6083,
|
| 1564 |
+
"step": 23000
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.86,
|
| 1568 |
+
"eval_loss": 0.6180054545402527,
|
| 1569 |
+
"eval_runtime": 1278.5639,
|
| 1570 |
+
"eval_samples_per_second": 0.424,
|
| 1571 |
+
"eval_steps_per_second": 0.424,
|
| 1572 |
+
"step": 23000
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 0.86,
|
| 1576 |
+
"learning_rate": 0.0001620327094880844,
|
| 1577 |
+
"loss": 0.5795,
|
| 1578 |
+
"step": 23100
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.86,
|
| 1582 |
+
"learning_rate": 0.00016172596075378893,
|
| 1583 |
+
"loss": 0.6025,
|
| 1584 |
+
"step": 23200
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.87,
|
| 1588 |
+
"learning_rate": 0.00016141827081809075,
|
| 1589 |
+
"loss": 0.5669,
|
| 1590 |
+
"step": 23300
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 0.87,
|
| 1594 |
+
"learning_rate": 0.00016110964437266568,
|
| 1595 |
+
"loss": 0.6172,
|
| 1596 |
+
"step": 23400
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 0.88,
|
| 1600 |
+
"learning_rate": 0.00016080008612346955,
|
| 1601 |
+
"loss": 0.5899,
|
| 1602 |
+
"step": 23500
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 0.88,
|
| 1606 |
+
"learning_rate": 0.00016048960079066636,
|
| 1607 |
+
"loss": 0.5889,
|
| 1608 |
+
"step": 23600
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 0.88,
|
| 1612 |
+
"learning_rate": 0.00016017819310855632,
|
| 1613 |
+
"loss": 0.5893,
|
| 1614 |
+
"step": 23700
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 0.89,
|
| 1618 |
+
"learning_rate": 0.00015986586782550376,
|
| 1619 |
+
"loss": 0.6363,
|
| 1620 |
+
"step": 23800
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.89,
|
| 1624 |
+
"learning_rate": 0.00015955262970386458,
|
| 1625 |
+
"loss": 0.5876,
|
| 1626 |
+
"step": 23900
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.89,
|
| 1630 |
+
"learning_rate": 0.00015923848351991372,
|
| 1631 |
+
"loss": 0.5964,
|
| 1632 |
+
"step": 24000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 0.89,
|
| 1636 |
+
"eval_loss": 0.6122664213180542,
|
| 1637 |
+
"eval_runtime": 1255.2341,
|
| 1638 |
+
"eval_samples_per_second": 0.432,
|
| 1639 |
+
"eval_steps_per_second": 0.432,
|
| 1640 |
+
"step": 24000
|
| 1641 |
}
|
| 1642 |
],
|
| 1643 |
"max_steps": 80463,
|
| 1644 |
"num_train_epochs": 3,
|
| 1645 |
+
"total_flos": 6.727787581059072e+18,
|
| 1646 |
"trial_name": null,
|
| 1647 |
"trial_params": null
|
| 1648 |
}
|
{checkpoint-20000 → checkpoint-24000}/training_args.bin
RENAMED
|
File without changes
|