aissatoubalde
commited on
Commit
•
7f23800
1
Parent(s):
e4aaddd
Training in progress, epoch 2
Browse files- adapter_model.safetensors +1 -1
- phi-2-custom-2.ipynb +346 -2
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 73433680
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:45fc3566b190817dd53b9455656ddd154dabc1e23233fb6f6bd57f14e54334f2
|
3 |
size 73433680
|
phi-2-custom-2.ipynb
CHANGED
@@ -1033,8 +1033,8 @@
|
|
1033 |
"\n",
|
1034 |
" <div>\n",
|
1035 |
" \n",
|
1036 |
-
" <progress value='
|
1037 |
-
" [
|
1038 |
" </div>\n",
|
1039 |
" <table border=\"1\" class=\"dataframe\">\n",
|
1040 |
" <thead>\n",
|
@@ -1332,6 +1332,350 @@
|
|
1332 |
" <td>3600</td>\n",
|
1333 |
" <td>1.090700</td>\n",
|
1334 |
" </tr>\n",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1335 |
" </tbody>\n",
|
1336 |
"</table><p>"
|
1337 |
],
|
|
|
1033 |
"\n",
|
1034 |
" <div>\n",
|
1035 |
" \n",
|
1036 |
+
" <progress value='7947' max='20000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
|
1037 |
+
" [ 7947/20000 29:33 < 44:51, 4.48 it/s, Epoch 1.94/5]\n",
|
1038 |
" </div>\n",
|
1039 |
" <table border=\"1\" class=\"dataframe\">\n",
|
1040 |
" <thead>\n",
|
|
|
1332 |
" <td>3600</td>\n",
|
1333 |
" <td>1.090700</td>\n",
|
1334 |
" </tr>\n",
|
1335 |
+
" <tr>\n",
|
1336 |
+
" <td>3650</td>\n",
|
1337 |
+
" <td>1.040200</td>\n",
|
1338 |
+
" </tr>\n",
|
1339 |
+
" <tr>\n",
|
1340 |
+
" <td>3700</td>\n",
|
1341 |
+
" <td>1.054600</td>\n",
|
1342 |
+
" </tr>\n",
|
1343 |
+
" <tr>\n",
|
1344 |
+
" <td>3750</td>\n",
|
1345 |
+
" <td>1.079500</td>\n",
|
1346 |
+
" </tr>\n",
|
1347 |
+
" <tr>\n",
|
1348 |
+
" <td>3800</td>\n",
|
1349 |
+
" <td>1.034800</td>\n",
|
1350 |
+
" </tr>\n",
|
1351 |
+
" <tr>\n",
|
1352 |
+
" <td>3850</td>\n",
|
1353 |
+
" <td>1.069000</td>\n",
|
1354 |
+
" </tr>\n",
|
1355 |
+
" <tr>\n",
|
1356 |
+
" <td>3900</td>\n",
|
1357 |
+
" <td>0.980800</td>\n",
|
1358 |
+
" </tr>\n",
|
1359 |
+
" <tr>\n",
|
1360 |
+
" <td>3950</td>\n",
|
1361 |
+
" <td>1.068200</td>\n",
|
1362 |
+
" </tr>\n",
|
1363 |
+
" <tr>\n",
|
1364 |
+
" <td>4000</td>\n",
|
1365 |
+
" <td>1.077300</td>\n",
|
1366 |
+
" </tr>\n",
|
1367 |
+
" <tr>\n",
|
1368 |
+
" <td>4050</td>\n",
|
1369 |
+
" <td>1.044200</td>\n",
|
1370 |
+
" </tr>\n",
|
1371 |
+
" <tr>\n",
|
1372 |
+
" <td>4100</td>\n",
|
1373 |
+
" <td>1.046300</td>\n",
|
1374 |
+
" </tr>\n",
|
1375 |
+
" <tr>\n",
|
1376 |
+
" <td>4150</td>\n",
|
1377 |
+
" <td>1.008200</td>\n",
|
1378 |
+
" </tr>\n",
|
1379 |
+
" <tr>\n",
|
1380 |
+
" <td>4200</td>\n",
|
1381 |
+
" <td>0.974100</td>\n",
|
1382 |
+
" </tr>\n",
|
1383 |
+
" <tr>\n",
|
1384 |
+
" <td>4250</td>\n",
|
1385 |
+
" <td>0.963200</td>\n",
|
1386 |
+
" </tr>\n",
|
1387 |
+
" <tr>\n",
|
1388 |
+
" <td>4300</td>\n",
|
1389 |
+
" <td>0.965000</td>\n",
|
1390 |
+
" </tr>\n",
|
1391 |
+
" <tr>\n",
|
1392 |
+
" <td>4350</td>\n",
|
1393 |
+
" <td>1.001900</td>\n",
|
1394 |
+
" </tr>\n",
|
1395 |
+
" <tr>\n",
|
1396 |
+
" <td>4400</td>\n",
|
1397 |
+
" <td>0.999400</td>\n",
|
1398 |
+
" </tr>\n",
|
1399 |
+
" <tr>\n",
|
1400 |
+
" <td>4450</td>\n",
|
1401 |
+
" <td>0.948700</td>\n",
|
1402 |
+
" </tr>\n",
|
1403 |
+
" <tr>\n",
|
1404 |
+
" <td>4500</td>\n",
|
1405 |
+
" <td>0.978800</td>\n",
|
1406 |
+
" </tr>\n",
|
1407 |
+
" <tr>\n",
|
1408 |
+
" <td>4550</td>\n",
|
1409 |
+
" <td>0.954600</td>\n",
|
1410 |
+
" </tr>\n",
|
1411 |
+
" <tr>\n",
|
1412 |
+
" <td>4600</td>\n",
|
1413 |
+
" <td>0.966300</td>\n",
|
1414 |
+
" </tr>\n",
|
1415 |
+
" <tr>\n",
|
1416 |
+
" <td>4650</td>\n",
|
1417 |
+
" <td>0.968500</td>\n",
|
1418 |
+
" </tr>\n",
|
1419 |
+
" <tr>\n",
|
1420 |
+
" <td>4700</td>\n",
|
1421 |
+
" <td>0.903800</td>\n",
|
1422 |
+
" </tr>\n",
|
1423 |
+
" <tr>\n",
|
1424 |
+
" <td>4750</td>\n",
|
1425 |
+
" <td>0.959700</td>\n",
|
1426 |
+
" </tr>\n",
|
1427 |
+
" <tr>\n",
|
1428 |
+
" <td>4800</td>\n",
|
1429 |
+
" <td>1.015000</td>\n",
|
1430 |
+
" </tr>\n",
|
1431 |
+
" <tr>\n",
|
1432 |
+
" <td>4850</td>\n",
|
1433 |
+
" <td>0.936600</td>\n",
|
1434 |
+
" </tr>\n",
|
1435 |
+
" <tr>\n",
|
1436 |
+
" <td>4900</td>\n",
|
1437 |
+
" <td>0.979400</td>\n",
|
1438 |
+
" </tr>\n",
|
1439 |
+
" <tr>\n",
|
1440 |
+
" <td>4950</td>\n",
|
1441 |
+
" <td>0.970100</td>\n",
|
1442 |
+
" </tr>\n",
|
1443 |
+
" <tr>\n",
|
1444 |
+
" <td>5000</td>\n",
|
1445 |
+
" <td>0.964700</td>\n",
|
1446 |
+
" </tr>\n",
|
1447 |
+
" <tr>\n",
|
1448 |
+
" <td>5050</td>\n",
|
1449 |
+
" <td>0.941400</td>\n",
|
1450 |
+
" </tr>\n",
|
1451 |
+
" <tr>\n",
|
1452 |
+
" <td>5100</td>\n",
|
1453 |
+
" <td>0.892600</td>\n",
|
1454 |
+
" </tr>\n",
|
1455 |
+
" <tr>\n",
|
1456 |
+
" <td>5150</td>\n",
|
1457 |
+
" <td>0.942300</td>\n",
|
1458 |
+
" </tr>\n",
|
1459 |
+
" <tr>\n",
|
1460 |
+
" <td>5200</td>\n",
|
1461 |
+
" <td>0.995600</td>\n",
|
1462 |
+
" </tr>\n",
|
1463 |
+
" <tr>\n",
|
1464 |
+
" <td>5250</td>\n",
|
1465 |
+
" <td>0.957400</td>\n",
|
1466 |
+
" </tr>\n",
|
1467 |
+
" <tr>\n",
|
1468 |
+
" <td>5300</td>\n",
|
1469 |
+
" <td>0.996400</td>\n",
|
1470 |
+
" </tr>\n",
|
1471 |
+
" <tr>\n",
|
1472 |
+
" <td>5350</td>\n",
|
1473 |
+
" <td>0.913700</td>\n",
|
1474 |
+
" </tr>\n",
|
1475 |
+
" <tr>\n",
|
1476 |
+
" <td>5400</td>\n",
|
1477 |
+
" <td>1.001200</td>\n",
|
1478 |
+
" </tr>\n",
|
1479 |
+
" <tr>\n",
|
1480 |
+
" <td>5450</td>\n",
|
1481 |
+
" <td>0.984900</td>\n",
|
1482 |
+
" </tr>\n",
|
1483 |
+
" <tr>\n",
|
1484 |
+
" <td>5500</td>\n",
|
1485 |
+
" <td>1.006300</td>\n",
|
1486 |
+
" </tr>\n",
|
1487 |
+
" <tr>\n",
|
1488 |
+
" <td>5550</td>\n",
|
1489 |
+
" <td>0.965200</td>\n",
|
1490 |
+
" </tr>\n",
|
1491 |
+
" <tr>\n",
|
1492 |
+
" <td>5600</td>\n",
|
1493 |
+
" <td>0.977800</td>\n",
|
1494 |
+
" </tr>\n",
|
1495 |
+
" <tr>\n",
|
1496 |
+
" <td>5650</td>\n",
|
1497 |
+
" <td>0.972400</td>\n",
|
1498 |
+
" </tr>\n",
|
1499 |
+
" <tr>\n",
|
1500 |
+
" <td>5700</td>\n",
|
1501 |
+
" <td>0.922800</td>\n",
|
1502 |
+
" </tr>\n",
|
1503 |
+
" <tr>\n",
|
1504 |
+
" <td>5750</td>\n",
|
1505 |
+
" <td>0.943600</td>\n",
|
1506 |
+
" </tr>\n",
|
1507 |
+
" <tr>\n",
|
1508 |
+
" <td>5800</td>\n",
|
1509 |
+
" <td>0.931300</td>\n",
|
1510 |
+
" </tr>\n",
|
1511 |
+
" <tr>\n",
|
1512 |
+
" <td>5850</td>\n",
|
1513 |
+
" <td>0.968300</td>\n",
|
1514 |
+
" </tr>\n",
|
1515 |
+
" <tr>\n",
|
1516 |
+
" <td>5900</td>\n",
|
1517 |
+
" <td>0.972600</td>\n",
|
1518 |
+
" </tr>\n",
|
1519 |
+
" <tr>\n",
|
1520 |
+
" <td>5950</td>\n",
|
1521 |
+
" <td>0.952300</td>\n",
|
1522 |
+
" </tr>\n",
|
1523 |
+
" <tr>\n",
|
1524 |
+
" <td>6000</td>\n",
|
1525 |
+
" <td>0.956700</td>\n",
|
1526 |
+
" </tr>\n",
|
1527 |
+
" <tr>\n",
|
1528 |
+
" <td>6050</td>\n",
|
1529 |
+
" <td>0.921800</td>\n",
|
1530 |
+
" </tr>\n",
|
1531 |
+
" <tr>\n",
|
1532 |
+
" <td>6100</td>\n",
|
1533 |
+
" <td>0.964500</td>\n",
|
1534 |
+
" </tr>\n",
|
1535 |
+
" <tr>\n",
|
1536 |
+
" <td>6150</td>\n",
|
1537 |
+
" <td>0.955100</td>\n",
|
1538 |
+
" </tr>\n",
|
1539 |
+
" <tr>\n",
|
1540 |
+
" <td>6200</td>\n",
|
1541 |
+
" <td>0.990300</td>\n",
|
1542 |
+
" </tr>\n",
|
1543 |
+
" <tr>\n",
|
1544 |
+
" <td>6250</td>\n",
|
1545 |
+
" <td>0.975900</td>\n",
|
1546 |
+
" </tr>\n",
|
1547 |
+
" <tr>\n",
|
1548 |
+
" <td>6300</td>\n",
|
1549 |
+
" <td>0.956100</td>\n",
|
1550 |
+
" </tr>\n",
|
1551 |
+
" <tr>\n",
|
1552 |
+
" <td>6350</td>\n",
|
1553 |
+
" <td>0.952400</td>\n",
|
1554 |
+
" </tr>\n",
|
1555 |
+
" <tr>\n",
|
1556 |
+
" <td>6400</td>\n",
|
1557 |
+
" <td>0.908500</td>\n",
|
1558 |
+
" </tr>\n",
|
1559 |
+
" <tr>\n",
|
1560 |
+
" <td>6450</td>\n",
|
1561 |
+
" <td>0.955700</td>\n",
|
1562 |
+
" </tr>\n",
|
1563 |
+
" <tr>\n",
|
1564 |
+
" <td>6500</td>\n",
|
1565 |
+
" <td>0.925000</td>\n",
|
1566 |
+
" </tr>\n",
|
1567 |
+
" <tr>\n",
|
1568 |
+
" <td>6550</td>\n",
|
1569 |
+
" <td>1.025000</td>\n",
|
1570 |
+
" </tr>\n",
|
1571 |
+
" <tr>\n",
|
1572 |
+
" <td>6600</td>\n",
|
1573 |
+
" <td>0.983300</td>\n",
|
1574 |
+
" </tr>\n",
|
1575 |
+
" <tr>\n",
|
1576 |
+
" <td>6650</td>\n",
|
1577 |
+
" <td>0.989700</td>\n",
|
1578 |
+
" </tr>\n",
|
1579 |
+
" <tr>\n",
|
1580 |
+
" <td>6700</td>\n",
|
1581 |
+
" <td>0.994500</td>\n",
|
1582 |
+
" </tr>\n",
|
1583 |
+
" <tr>\n",
|
1584 |
+
" <td>6750</td>\n",
|
1585 |
+
" <td>0.997600</td>\n",
|
1586 |
+
" </tr>\n",
|
1587 |
+
" <tr>\n",
|
1588 |
+
" <td>6800</td>\n",
|
1589 |
+
" <td>1.062700</td>\n",
|
1590 |
+
" </tr>\n",
|
1591 |
+
" <tr>\n",
|
1592 |
+
" <td>6850</td>\n",
|
1593 |
+
" <td>0.982500</td>\n",
|
1594 |
+
" </tr>\n",
|
1595 |
+
" <tr>\n",
|
1596 |
+
" <td>6900</td>\n",
|
1597 |
+
" <td>0.960900</td>\n",
|
1598 |
+
" </tr>\n",
|
1599 |
+
" <tr>\n",
|
1600 |
+
" <td>6950</td>\n",
|
1601 |
+
" <td>0.914900</td>\n",
|
1602 |
+
" </tr>\n",
|
1603 |
+
" <tr>\n",
|
1604 |
+
" <td>7000</td>\n",
|
1605 |
+
" <td>0.956200</td>\n",
|
1606 |
+
" </tr>\n",
|
1607 |
+
" <tr>\n",
|
1608 |
+
" <td>7050</td>\n",
|
1609 |
+
" <td>0.880100</td>\n",
|
1610 |
+
" </tr>\n",
|
1611 |
+
" <tr>\n",
|
1612 |
+
" <td>7100</td>\n",
|
1613 |
+
" <td>0.954500</td>\n",
|
1614 |
+
" </tr>\n",
|
1615 |
+
" <tr>\n",
|
1616 |
+
" <td>7150</td>\n",
|
1617 |
+
" <td>0.911300</td>\n",
|
1618 |
+
" </tr>\n",
|
1619 |
+
" <tr>\n",
|
1620 |
+
" <td>7200</td>\n",
|
1621 |
+
" <td>0.939300</td>\n",
|
1622 |
+
" </tr>\n",
|
1623 |
+
" <tr>\n",
|
1624 |
+
" <td>7250</td>\n",
|
1625 |
+
" <td>0.944700</td>\n",
|
1626 |
+
" </tr>\n",
|
1627 |
+
" <tr>\n",
|
1628 |
+
" <td>7300</td>\n",
|
1629 |
+
" <td>0.915300</td>\n",
|
1630 |
+
" </tr>\n",
|
1631 |
+
" <tr>\n",
|
1632 |
+
" <td>7350</td>\n",
|
1633 |
+
" <td>0.919300</td>\n",
|
1634 |
+
" </tr>\n",
|
1635 |
+
" <tr>\n",
|
1636 |
+
" <td>7400</td>\n",
|
1637 |
+
" <td>0.909000</td>\n",
|
1638 |
+
" </tr>\n",
|
1639 |
+
" <tr>\n",
|
1640 |
+
" <td>7450</td>\n",
|
1641 |
+
" <td>0.950200</td>\n",
|
1642 |
+
" </tr>\n",
|
1643 |
+
" <tr>\n",
|
1644 |
+
" <td>7500</td>\n",
|
1645 |
+
" <td>0.958200</td>\n",
|
1646 |
+
" </tr>\n",
|
1647 |
+
" <tr>\n",
|
1648 |
+
" <td>7550</td>\n",
|
1649 |
+
" <td>0.906400</td>\n",
|
1650 |
+
" </tr>\n",
|
1651 |
+
" <tr>\n",
|
1652 |
+
" <td>7600</td>\n",
|
1653 |
+
" <td>0.929400</td>\n",
|
1654 |
+
" </tr>\n",
|
1655 |
+
" <tr>\n",
|
1656 |
+
" <td>7650</td>\n",
|
1657 |
+
" <td>1.000700</td>\n",
|
1658 |
+
" </tr>\n",
|
1659 |
+
" <tr>\n",
|
1660 |
+
" <td>7700</td>\n",
|
1661 |
+
" <td>0.952400</td>\n",
|
1662 |
+
" </tr>\n",
|
1663 |
+
" <tr>\n",
|
1664 |
+
" <td>7750</td>\n",
|
1665 |
+
" <td>0.938000</td>\n",
|
1666 |
+
" </tr>\n",
|
1667 |
+
" <tr>\n",
|
1668 |
+
" <td>7800</td>\n",
|
1669 |
+
" <td>0.932500</td>\n",
|
1670 |
+
" </tr>\n",
|
1671 |
+
" <tr>\n",
|
1672 |
+
" <td>7850</td>\n",
|
1673 |
+
" <td>0.952100</td>\n",
|
1674 |
+
" </tr>\n",
|
1675 |
+
" <tr>\n",
|
1676 |
+
" <td>7900</td>\n",
|
1677 |
+
" <td>0.931400</td>\n",
|
1678 |
+
" </tr>\n",
|
1679 |
" </tbody>\n",
|
1680 |
"</table><p>"
|
1681 |
],
|