Training in progress, step 250, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 671149168
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e5e70313e832a97b4a5f909837ad9943129a92cc5051d6a727def5672f1a649
|
| 3 |
size 671149168
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1342555602
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2d53e875526cc7f45b9ee008df430f17f426411218e934f51ade9ef8b575fcb
|
| 3 |
size 1342555602
|
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 sha256:3055d414afa45ec35d956573da41804fc6e7dcd64cd4972d9d8a03f439aed732
|
| 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 sha256:0cc9840a370c6adb27f4a6f97aa77bff7e496e9a0cca1d81b87512770179ee98
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "miner_id_24/checkpoint-
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 50,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1447,6 +1447,364 @@
|
|
| 1447 |
"eval_samples_per_second": 5.533,
|
| 1448 |
"eval_steps_per_second": 2.767,
|
| 1449 |
"step": 200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1450 |
}
|
| 1451 |
],
|
| 1452 |
"logging_steps": 1,
|
|
@@ -1475,7 +1833,7 @@
|
|
| 1475 |
"attributes": {}
|
| 1476 |
}
|
| 1477 |
},
|
| 1478 |
-
"total_flos":
|
| 1479 |
"train_batch_size": 8,
|
| 1480 |
"trial_name": null,
|
| 1481 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5557529926300049,
|
| 3 |
+
"best_model_checkpoint": "miner_id_24/checkpoint-250",
|
| 4 |
+
"epoch": 0.05714612263557917,
|
| 5 |
"eval_steps": 50,
|
| 6 |
+
"global_step": 250,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1447 |
"eval_samples_per_second": 5.533,
|
| 1448 |
"eval_steps_per_second": 2.767,
|
| 1449 |
"step": 200
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.04594548259900566,
|
| 1453 |
+
"grad_norm": 10.6249418258667,
|
| 1454 |
+
"learning_rate": 2.9663167846209998e-05,
|
| 1455 |
+
"loss": 3.5265,
|
| 1456 |
+
"step": 201
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 0.04617406708954797,
|
| 1460 |
+
"grad_norm": 8.47287368774414,
|
| 1461 |
+
"learning_rate": 2.9133077275909108e-05,
|
| 1462 |
+
"loss": 3.4879,
|
| 1463 |
+
"step": 202
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 0.04640265158009029,
|
| 1467 |
+
"grad_norm": 7.781945705413818,
|
| 1468 |
+
"learning_rate": 2.86058117529173e-05,
|
| 1469 |
+
"loss": 3.0601,
|
| 1470 |
+
"step": 203
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 0.04663123607063261,
|
| 1474 |
+
"grad_norm": 7.176991939544678,
|
| 1475 |
+
"learning_rate": 2.8081442660546125e-05,
|
| 1476 |
+
"loss": 2.6794,
|
| 1477 |
+
"step": 204
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 0.04685982056117492,
|
| 1481 |
+
"grad_norm": 6.956686496734619,
|
| 1482 |
+
"learning_rate": 2.7560040989976892e-05,
|
| 1483 |
+
"loss": 2.3741,
|
| 1484 |
+
"step": 205
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 0.04708840505171724,
|
| 1488 |
+
"grad_norm": 7.541036605834961,
|
| 1489 |
+
"learning_rate": 2.7041677330649407e-05,
|
| 1490 |
+
"loss": 2.4602,
|
| 1491 |
+
"step": 206
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 0.04731698954225956,
|
| 1495 |
+
"grad_norm": 26.17896270751953,
|
| 1496 |
+
"learning_rate": 2.6526421860705473e-05,
|
| 1497 |
+
"loss": 1.7172,
|
| 1498 |
+
"step": 207
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 0.04754557403280187,
|
| 1502 |
+
"grad_norm": 44.718421936035156,
|
| 1503 |
+
"learning_rate": 2.6014344337487707e-05,
|
| 1504 |
+
"loss": 1.8271,
|
| 1505 |
+
"step": 208
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 0.04777415852334419,
|
| 1509 |
+
"grad_norm": 23.585582733154297,
|
| 1510 |
+
"learning_rate": 2.5505514088095655e-05,
|
| 1511 |
+
"loss": 1.1668,
|
| 1512 |
+
"step": 209
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 0.04800274301388651,
|
| 1516 |
+
"grad_norm": 18.206127166748047,
|
| 1517 |
+
"learning_rate": 2.500000000000001e-05,
|
| 1518 |
+
"loss": 0.7469,
|
| 1519 |
+
"step": 210
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 0.04823132750442882,
|
| 1523 |
+
"grad_norm": 4.393265724182129,
|
| 1524 |
+
"learning_rate": 2.4497870511716235e-05,
|
| 1525 |
+
"loss": 0.1077,
|
| 1526 |
+
"step": 211
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 0.04845991199497114,
|
| 1530 |
+
"grad_norm": 4.106040954589844,
|
| 1531 |
+
"learning_rate": 2.399919360353923e-05,
|
| 1532 |
+
"loss": 0.2907,
|
| 1533 |
+
"step": 212
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 0.04868849648551346,
|
| 1537 |
+
"grad_norm": 13.828383445739746,
|
| 1538 |
+
"learning_rate": 2.350403678833976e-05,
|
| 1539 |
+
"loss": 0.8737,
|
| 1540 |
+
"step": 213
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 0.04891708097605577,
|
| 1544 |
+
"grad_norm": 59.581298828125,
|
| 1545 |
+
"learning_rate": 2.3012467102424373e-05,
|
| 1546 |
+
"loss": 4.8994,
|
| 1547 |
+
"step": 214
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 0.04914566546659809,
|
| 1551 |
+
"grad_norm": 52.219642639160156,
|
| 1552 |
+
"learning_rate": 2.25245510964597e-05,
|
| 1553 |
+
"loss": 4.3086,
|
| 1554 |
+
"step": 215
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 0.04937424995714041,
|
| 1558 |
+
"grad_norm": 57.01164627075195,
|
| 1559 |
+
"learning_rate": 2.2040354826462668e-05,
|
| 1560 |
+
"loss": 3.0226,
|
| 1561 |
+
"step": 216
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 0.04960283444768272,
|
| 1565 |
+
"grad_norm": 35.1441764831543,
|
| 1566 |
+
"learning_rate": 2.1559943844857422e-05,
|
| 1567 |
+
"loss": 1.8326,
|
| 1568 |
+
"step": 217
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 0.04983141893822504,
|
| 1572 |
+
"grad_norm": 23.194089889526367,
|
| 1573 |
+
"learning_rate": 2.1083383191600674e-05,
|
| 1574 |
+
"loss": 0.9515,
|
| 1575 |
+
"step": 218
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 0.05006000342876736,
|
| 1579 |
+
"grad_norm": 20.370725631713867,
|
| 1580 |
+
"learning_rate": 2.061073738537635e-05,
|
| 1581 |
+
"loss": 2.2338,
|
| 1582 |
+
"step": 219
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 0.05028858791930967,
|
| 1586 |
+
"grad_norm": 31.7033634185791,
|
| 1587 |
+
"learning_rate": 2.0142070414860704e-05,
|
| 1588 |
+
"loss": 2.838,
|
| 1589 |
+
"step": 220
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 0.05051717240985199,
|
| 1593 |
+
"grad_norm": 30.212299346923828,
|
| 1594 |
+
"learning_rate": 1.9677445730059346e-05,
|
| 1595 |
+
"loss": 3.5437,
|
| 1596 |
+
"step": 221
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 0.05074575690039431,
|
| 1600 |
+
"grad_norm": 24.48371124267578,
|
| 1601 |
+
"learning_rate": 1.9216926233717085e-05,
|
| 1602 |
+
"loss": 2.1516,
|
| 1603 |
+
"step": 222
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 0.05097434139093662,
|
| 1607 |
+
"grad_norm": 26.28635597229004,
|
| 1608 |
+
"learning_rate": 1.8760574272802e-05,
|
| 1609 |
+
"loss": 1.8107,
|
| 1610 |
+
"step": 223
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 0.05120292588147894,
|
| 1614 |
+
"grad_norm": 64.8018569946289,
|
| 1615 |
+
"learning_rate": 1.8308451630064484e-05,
|
| 1616 |
+
"loss": 3.8766,
|
| 1617 |
+
"step": 224
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 0.05143151037202126,
|
| 1621 |
+
"grad_norm": 79.22872161865234,
|
| 1622 |
+
"learning_rate": 1.7860619515673033e-05,
|
| 1623 |
+
"loss": 5.5396,
|
| 1624 |
+
"step": 225
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 0.05166009486256357,
|
| 1628 |
+
"grad_norm": 81.44451904296875,
|
| 1629 |
+
"learning_rate": 1.7417138558927244e-05,
|
| 1630 |
+
"loss": 5.3564,
|
| 1631 |
+
"step": 226
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 0.05188867935310589,
|
| 1635 |
+
"grad_norm": 74.99747467041016,
|
| 1636 |
+
"learning_rate": 1.6978068800049624e-05,
|
| 1637 |
+
"loss": 4.2999,
|
| 1638 |
+
"step": 227
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 0.05211726384364821,
|
| 1642 |
+
"grad_norm": 30.039758682250977,
|
| 1643 |
+
"learning_rate": 1.6543469682057106e-05,
|
| 1644 |
+
"loss": 4.5715,
|
| 1645 |
+
"step": 228
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 0.05234584833419052,
|
| 1649 |
+
"grad_norm": 25.890417098999023,
|
| 1650 |
+
"learning_rate": 1.611340004271339e-05,
|
| 1651 |
+
"loss": 4.2147,
|
| 1652 |
+
"step": 229
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 0.05257443282473284,
|
| 1656 |
+
"grad_norm": 24.455820083618164,
|
| 1657 |
+
"learning_rate": 1.5687918106563326e-05,
|
| 1658 |
+
"loss": 3.6963,
|
| 1659 |
+
"step": 230
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 0.05280301731527516,
|
| 1663 |
+
"grad_norm": 13.776949882507324,
|
| 1664 |
+
"learning_rate": 1.526708147705013e-05,
|
| 1665 |
+
"loss": 2.5523,
|
| 1666 |
+
"step": 231
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 0.05303160180581747,
|
| 1670 |
+
"grad_norm": 14.489748001098633,
|
| 1671 |
+
"learning_rate": 1.4850947128716913e-05,
|
| 1672 |
+
"loss": 2.8481,
|
| 1673 |
+
"step": 232
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 0.05326018629635979,
|
| 1677 |
+
"grad_norm": 14.851402282714844,
|
| 1678 |
+
"learning_rate": 1.4439571399493146e-05,
|
| 1679 |
+
"loss": 2.9429,
|
| 1680 |
+
"step": 233
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 0.05348877078690211,
|
| 1684 |
+
"grad_norm": 14.959819793701172,
|
| 1685 |
+
"learning_rate": 1.4033009983067452e-05,
|
| 1686 |
+
"loss": 2.7784,
|
| 1687 |
+
"step": 234
|
| 1688 |
+
},
|
| 1689 |
+
{
|
| 1690 |
+
"epoch": 0.05371735527744442,
|
| 1691 |
+
"grad_norm": 10.266538619995117,
|
| 1692 |
+
"learning_rate": 1.3631317921347563e-05,
|
| 1693 |
+
"loss": 2.1112,
|
| 1694 |
+
"step": 235
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 0.05394593976798674,
|
| 1698 |
+
"grad_norm": 13.498571395874023,
|
| 1699 |
+
"learning_rate": 1.3234549597008571e-05,
|
| 1700 |
+
"loss": 2.61,
|
| 1701 |
+
"step": 236
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 0.05417452425852906,
|
| 1705 |
+
"grad_norm": 13.794647216796875,
|
| 1706 |
+
"learning_rate": 1.2842758726130283e-05,
|
| 1707 |
+
"loss": 2.655,
|
| 1708 |
+
"step": 237
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 0.05440310874907137,
|
| 1712 |
+
"grad_norm": 12.700047492980957,
|
| 1713 |
+
"learning_rate": 1.245599835092504e-05,
|
| 1714 |
+
"loss": 2.1982,
|
| 1715 |
+
"step": 238
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 0.05463169323961369,
|
| 1719 |
+
"grad_norm": 11.33856201171875,
|
| 1720 |
+
"learning_rate": 1.2074320832556556e-05,
|
| 1721 |
+
"loss": 1.8852,
|
| 1722 |
+
"step": 239
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 0.05486027773015601,
|
| 1726 |
+
"grad_norm": 12.628767013549805,
|
| 1727 |
+
"learning_rate": 1.1697777844051105e-05,
|
| 1728 |
+
"loss": 2.6147,
|
| 1729 |
+
"step": 240
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 0.05508886222069832,
|
| 1733 |
+
"grad_norm": 12.457136154174805,
|
| 1734 |
+
"learning_rate": 1.132642036330181e-05,
|
| 1735 |
+
"loss": 2.2797,
|
| 1736 |
+
"step": 241
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 0.05531744671124064,
|
| 1740 |
+
"grad_norm": 16.78705406188965,
|
| 1741 |
+
"learning_rate": 1.096029866616704e-05,
|
| 1742 |
+
"loss": 2.1772,
|
| 1743 |
+
"step": 242
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 0.05554603120178296,
|
| 1747 |
+
"grad_norm": 20.40379524230957,
|
| 1748 |
+
"learning_rate": 1.0599462319663905e-05,
|
| 1749 |
+
"loss": 4.4865,
|
| 1750 |
+
"step": 243
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 0.05577461569232527,
|
| 1754 |
+
"grad_norm": 47.940460205078125,
|
| 1755 |
+
"learning_rate": 1.0243960175257606e-05,
|
| 1756 |
+
"loss": 4.1248,
|
| 1757 |
+
"step": 244
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 0.05600320018286759,
|
| 1761 |
+
"grad_norm": 20.962236404418945,
|
| 1762 |
+
"learning_rate": 9.893840362247809e-06,
|
| 1763 |
+
"loss": 5.0609,
|
| 1764 |
+
"step": 245
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 0.05623178467340991,
|
| 1768 |
+
"grad_norm": 17.944730758666992,
|
| 1769 |
+
"learning_rate": 9.549150281252633e-06,
|
| 1770 |
+
"loss": 5.8087,
|
| 1771 |
+
"step": 246
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 0.05646036916395222,
|
| 1775 |
+
"grad_norm": 14.370197296142578,
|
| 1776 |
+
"learning_rate": 9.209936597791407e-06,
|
| 1777 |
+
"loss": 4.9449,
|
| 1778 |
+
"step": 247
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 0.05668895365449454,
|
| 1782 |
+
"grad_norm": 13.268957138061523,
|
| 1783 |
+
"learning_rate": 8.876245235966885e-06,
|
| 1784 |
+
"loss": 4.4975,
|
| 1785 |
+
"step": 248
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 0.05691753814503686,
|
| 1789 |
+
"grad_norm": 13.039083480834961,
|
| 1790 |
+
"learning_rate": 8.548121372247918e-06,
|
| 1791 |
+
"loss": 3.907,
|
| 1792 |
+
"step": 249
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 0.05714612263557917,
|
| 1796 |
+
"grad_norm": 14.2072172164917,
|
| 1797 |
+
"learning_rate": 8.225609429353187e-06,
|
| 1798 |
+
"loss": 3.7406,
|
| 1799 |
+
"step": 250
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 0.05714612263557917,
|
| 1803 |
+
"eval_loss": 0.5557529926300049,
|
| 1804 |
+
"eval_runtime": 1331.7964,
|
| 1805 |
+
"eval_samples_per_second": 5.532,
|
| 1806 |
+
"eval_steps_per_second": 2.766,
|
| 1807 |
+
"step": 250
|
| 1808 |
}
|
| 1809 |
],
|
| 1810 |
"logging_steps": 1,
|
|
|
|
| 1833 |
"attributes": {}
|
| 1834 |
}
|
| 1835 |
},
|
| 1836 |
+
"total_flos": 5.566582254961951e+17,
|
| 1837 |
"train_batch_size": 8,
|
| 1838 |
"trial_name": null,
|
| 1839 |
"trial_params": null
|