Sara Price commited on
Commit
2b0bd6f
·
verified ·
1 Parent(s): 0e7d91f

Training in progress, step 6000, checkpoint

Browse files
last-checkpoint/model-00001-of-00006.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:25d11fffa78fa3b1e0350cb2054050d5672ee1113ba331c76dee0263cf4c1e45
3
  size 4840658560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb6e91353faedd52915fdb5c1b028989cf8ef6ce8683c51c7c9c5909111bb250
3
  size 4840658560
last-checkpoint/model-00002-of-00006.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ee934ed0f652e02b746ae354f51700e812a5d4b3e9afea154a700ad1b43201e0
3
  size 4857206856
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:969a8978e79c04af71a49f0c52ad091887a3ff6f5be64d95cd4a7e1873dec3bc
3
  size 4857206856
last-checkpoint/model-00003-of-00006.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:575f83806cd9a96c01c873eadf7ff19af4a57bc98ae2419dd835647e52b98105
3
  size 4857206904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e06d9de3831bc3966a87f5ad87d1f7b96ef29da8e73e560428da9becfaa6a7c
3
  size 4857206904
last-checkpoint/model-00004-of-00006.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:25b8499dc4c76e9ab59b05d1240165777eb091ad4bab7a141b374a5ef8c0925b
3
  size 4857206904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d545e04fae96e70c7894d32b7a1f970fb78d7ecde7de182623cabef7f26e7cd
3
  size 4857206904
last-checkpoint/model-00005-of-00006.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b179697a42f951366ca90f3adbfe0509dc9d406ab1fa03fc8d7fc9667e5a2c83
3
  size 4857206904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6b7c2f97132858a2e269ff55e19269be5f80ced2708eccb55d9bb59e5800dc1
3
  size 4857206904
last-checkpoint/model-00006-of-00006.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:798b9f30a23ef90a339a39e5fc032abdfa44c3917f4314068f8bbd5875728f11
3
  size 2684734256
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdac51d0613406ce47ce21fd846015c293307ce35adfa3b92b9402eb2811925e
3
  size 2684734256
last-checkpoint/trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 3.2,
5
  "eval_steps": 50,
6
- "global_step": 4000,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -1207,6 +1207,606 @@
1207
  "eval_samples_per_second": 46.932,
1208
  "eval_steps_per_second": 2.47,
1209
  "step": 4000
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1210
  }
1211
  ],
1212
  "logging_steps": 50,
@@ -1214,7 +1814,7 @@
1214
  "num_input_tokens_seen": 0,
1215
  "num_train_epochs": 10,
1216
  "save_steps": 2000,
1217
- "total_flos": 2.579446129062052e+17,
1218
  "train_batch_size": 4,
1219
  "trial_name": null,
1220
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 4.8,
5
  "eval_steps": 50,
6
+ "global_step": 6000,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
1207
  "eval_samples_per_second": 46.932,
1208
  "eval_steps_per_second": 2.47,
1209
  "step": 4000
1210
+ },
1211
+ {
1212
+ "epoch": 3.24,
1213
+ "grad_norm": 1.961307406425476,
1214
+ "learning_rate": 2.5643561048047815e-05,
1215
+ "loss": 0.1961,
1216
+ "step": 4050
1217
+ },
1218
+ {
1219
+ "epoch": 3.24,
1220
+ "eval_loss": 1.270025372505188,
1221
+ "eval_runtime": 2.5636,
1222
+ "eval_samples_per_second": 44.468,
1223
+ "eval_steps_per_second": 2.34,
1224
+ "step": 4050
1225
+ },
1226
+ {
1227
+ "epoch": 3.2800000000000002,
1228
+ "grad_norm": 1.6925843954086304,
1229
+ "learning_rate": 2.5494950107700482e-05,
1230
+ "loss": 0.2017,
1231
+ "step": 4100
1232
+ },
1233
+ {
1234
+ "epoch": 3.2800000000000002,
1235
+ "eval_loss": 1.2311433553695679,
1236
+ "eval_runtime": 2.6262,
1237
+ "eval_samples_per_second": 43.409,
1238
+ "eval_steps_per_second": 2.285,
1239
+ "step": 4100
1240
+ },
1241
+ {
1242
+ "epoch": 3.32,
1243
+ "grad_norm": 1.9287723302841187,
1244
+ "learning_rate": 2.5344293156035048e-05,
1245
+ "loss": 0.1986,
1246
+ "step": 4150
1247
+ },
1248
+ {
1249
+ "epoch": 3.32,
1250
+ "eval_loss": 1.2489361763000488,
1251
+ "eval_runtime": 2.5769,
1252
+ "eval_samples_per_second": 44.239,
1253
+ "eval_steps_per_second": 2.328,
1254
+ "step": 4150
1255
+ },
1256
+ {
1257
+ "epoch": 3.36,
1258
+ "grad_norm": 2.5327305793762207,
1259
+ "learning_rate": 2.519161956392275e-05,
1260
+ "loss": 0.2028,
1261
+ "step": 4200
1262
+ },
1263
+ {
1264
+ "epoch": 3.36,
1265
+ "eval_loss": 1.249537706375122,
1266
+ "eval_runtime": 2.4386,
1267
+ "eval_samples_per_second": 46.748,
1268
+ "eval_steps_per_second": 2.46,
1269
+ "step": 4200
1270
+ },
1271
+ {
1272
+ "epoch": 3.4,
1273
+ "grad_norm": 1.8875576257705688,
1274
+ "learning_rate": 2.5036959095382875e-05,
1275
+ "loss": 0.2151,
1276
+ "step": 4250
1277
+ },
1278
+ {
1279
+ "epoch": 3.4,
1280
+ "eval_loss": 1.2642710208892822,
1281
+ "eval_runtime": 2.4288,
1282
+ "eval_samples_per_second": 46.938,
1283
+ "eval_steps_per_second": 2.47,
1284
+ "step": 4250
1285
+ },
1286
+ {
1287
+ "epoch": 3.44,
1288
+ "grad_norm": 1.534134864807129,
1289
+ "learning_rate": 2.4880341901780205e-05,
1290
+ "loss": 0.2043,
1291
+ "step": 4300
1292
+ },
1293
+ {
1294
+ "epoch": 3.44,
1295
+ "eval_loss": 1.2751779556274414,
1296
+ "eval_runtime": 2.4239,
1297
+ "eval_samples_per_second": 47.031,
1298
+ "eval_steps_per_second": 2.475,
1299
+ "step": 4300
1300
+ },
1301
+ {
1302
+ "epoch": 3.48,
1303
+ "grad_norm": 1.740895390510559,
1304
+ "learning_rate": 2.4721798515946964e-05,
1305
+ "loss": 0.2146,
1306
+ "step": 4350
1307
+ },
1308
+ {
1309
+ "epoch": 3.48,
1310
+ "eval_loss": 1.2346069812774658,
1311
+ "eval_runtime": 2.4299,
1312
+ "eval_samples_per_second": 46.916,
1313
+ "eval_steps_per_second": 2.469,
1314
+ "step": 4350
1315
+ },
1316
+ {
1317
+ "epoch": 3.52,
1318
+ "grad_norm": 1.9260108470916748,
1319
+ "learning_rate": 2.4561359846230346e-05,
1320
+ "loss": 0.2088,
1321
+ "step": 4400
1322
+ },
1323
+ {
1324
+ "epoch": 3.52,
1325
+ "eval_loss": 1.2757267951965332,
1326
+ "eval_runtime": 2.4311,
1327
+ "eval_samples_per_second": 46.891,
1328
+ "eval_steps_per_second": 2.468,
1329
+ "step": 4400
1330
+ },
1331
+ {
1332
+ "epoch": 3.56,
1333
+ "grad_norm": 1.5608491897583008,
1334
+ "learning_rate": 2.439905717046691e-05,
1335
+ "loss": 0.2136,
1336
+ "step": 4450
1337
+ },
1338
+ {
1339
+ "epoch": 3.56,
1340
+ "eval_loss": 1.2627815008163452,
1341
+ "eval_runtime": 2.4322,
1342
+ "eval_samples_per_second": 46.871,
1343
+ "eval_steps_per_second": 2.467,
1344
+ "step": 4450
1345
+ },
1346
+ {
1347
+ "epoch": 3.6,
1348
+ "grad_norm": 2.380476713180542,
1349
+ "learning_rate": 2.4234922129884873e-05,
1350
+ "loss": 0.2139,
1351
+ "step": 4500
1352
+ },
1353
+ {
1354
+ "epoch": 3.6,
1355
+ "eval_loss": 1.283263921737671,
1356
+ "eval_runtime": 2.4366,
1357
+ "eval_samples_per_second": 46.787,
1358
+ "eval_steps_per_second": 2.462,
1359
+ "step": 4500
1360
+ },
1361
+ {
1362
+ "epoch": 3.64,
1363
+ "grad_norm": 1.572986125946045,
1364
+ "learning_rate": 2.4068986722935625e-05,
1365
+ "loss": 0.2108,
1366
+ "step": 4550
1367
+ },
1368
+ {
1369
+ "epoch": 3.64,
1370
+ "eval_loss": 1.2481166124343872,
1371
+ "eval_runtime": 2.4258,
1372
+ "eval_samples_per_second": 46.996,
1373
+ "eval_steps_per_second": 2.473,
1374
+ "step": 4550
1375
+ },
1376
+ {
1377
+ "epoch": 3.68,
1378
+ "grad_norm": 2.0618062019348145,
1379
+ "learning_rate": 2.3901283299055524e-05,
1380
+ "loss": 0.2076,
1381
+ "step": 4600
1382
+ },
1383
+ {
1384
+ "epoch": 3.68,
1385
+ "eval_loss": 1.2624186277389526,
1386
+ "eval_runtime": 2.4384,
1387
+ "eval_samples_per_second": 46.752,
1388
+ "eval_steps_per_second": 2.461,
1389
+ "step": 4600
1390
+ },
1391
+ {
1392
+ "epoch": 3.7199999999999998,
1393
+ "grad_norm": 1.5757406949996948,
1394
+ "learning_rate": 2.3731844552359342e-05,
1395
+ "loss": 0.2178,
1396
+ "step": 4650
1397
+ },
1398
+ {
1399
+ "epoch": 3.7199999999999998,
1400
+ "eval_loss": 1.2678072452545166,
1401
+ "eval_runtime": 2.4351,
1402
+ "eval_samples_per_second": 46.816,
1403
+ "eval_steps_per_second": 2.464,
1404
+ "step": 4650
1405
+ },
1406
+ {
1407
+ "epoch": 3.76,
1408
+ "grad_norm": 2.3872954845428467,
1409
+ "learning_rate": 2.356070351526648e-05,
1410
+ "loss": 0.2128,
1411
+ "step": 4700
1412
+ },
1413
+ {
1414
+ "epoch": 3.76,
1415
+ "eval_loss": 1.275978446006775,
1416
+ "eval_runtime": 2.4323,
1417
+ "eval_samples_per_second": 46.869,
1418
+ "eval_steps_per_second": 2.467,
1419
+ "step": 4700
1420
+ },
1421
+ {
1422
+ "epoch": 3.8,
1423
+ "grad_norm": 1.801849126815796,
1424
+ "learning_rate": 2.3387893552061202e-05,
1425
+ "loss": 0.209,
1426
+ "step": 4750
1427
+ },
1428
+ {
1429
+ "epoch": 3.8,
1430
+ "eval_loss": 1.2716574668884277,
1431
+ "eval_runtime": 2.4373,
1432
+ "eval_samples_per_second": 46.773,
1433
+ "eval_steps_per_second": 2.462,
1434
+ "step": 4750
1435
+ },
1436
+ {
1437
+ "epoch": 3.84,
1438
+ "grad_norm": 2.122264862060547,
1439
+ "learning_rate": 2.3213448352388256e-05,
1440
+ "loss": 0.2153,
1441
+ "step": 4800
1442
+ },
1443
+ {
1444
+ "epoch": 3.84,
1445
+ "eval_loss": 1.279623031616211,
1446
+ "eval_runtime": 2.438,
1447
+ "eval_samples_per_second": 46.759,
1448
+ "eval_steps_per_second": 2.461,
1449
+ "step": 4800
1450
+ },
1451
+ {
1452
+ "epoch": 3.88,
1453
+ "grad_norm": 2.083524465560913,
1454
+ "learning_rate": 2.303740192468495e-05,
1455
+ "loss": 0.2191,
1456
+ "step": 4850
1457
+ },
1458
+ {
1459
+ "epoch": 3.88,
1460
+ "eval_loss": 1.254883885383606,
1461
+ "eval_runtime": 2.4286,
1462
+ "eval_samples_per_second": 46.942,
1463
+ "eval_steps_per_second": 2.471,
1464
+ "step": 4850
1465
+ },
1466
+ {
1467
+ "epoch": 3.92,
1468
+ "grad_norm": 1.8724983930587769,
1469
+ "learning_rate": 2.285978858955119e-05,
1470
+ "loss": 0.212,
1471
+ "step": 4900
1472
+ },
1473
+ {
1474
+ "epoch": 3.92,
1475
+ "eval_loss": 1.2345080375671387,
1476
+ "eval_runtime": 2.4308,
1477
+ "eval_samples_per_second": 46.898,
1478
+ "eval_steps_per_second": 2.468,
1479
+ "step": 4900
1480
+ },
1481
+ {
1482
+ "epoch": 3.96,
1483
+ "grad_norm": 1.4495961666107178,
1484
+ "learning_rate": 2.2680642973058574e-05,
1485
+ "loss": 0.2168,
1486
+ "step": 4950
1487
+ },
1488
+ {
1489
+ "epoch": 3.96,
1490
+ "eval_loss": 1.2639931440353394,
1491
+ "eval_runtime": 2.4303,
1492
+ "eval_samples_per_second": 46.907,
1493
+ "eval_steps_per_second": 2.469,
1494
+ "step": 4950
1495
+ },
1496
+ {
1497
+ "epoch": 4.0,
1498
+ "grad_norm": 2.2413907051086426,
1499
+ "learning_rate": 2.25e-05,
1500
+ "loss": 0.2097,
1501
+ "step": 5000
1502
+ },
1503
+ {
1504
+ "epoch": 4.0,
1505
+ "eval_loss": 1.4285211563110352,
1506
+ "eval_runtime": 2.4274,
1507
+ "eval_samples_per_second": 46.963,
1508
+ "eval_steps_per_second": 2.472,
1509
+ "step": 5000
1510
+ },
1511
+ {
1512
+ "epoch": 4.04,
1513
+ "grad_norm": 1.1646087169647217,
1514
+ "learning_rate": 2.2317894887080992e-05,
1515
+ "loss": 0.1303,
1516
+ "step": 5050
1517
+ },
1518
+ {
1519
+ "epoch": 4.04,
1520
+ "eval_loss": 1.4936164617538452,
1521
+ "eval_runtime": 2.432,
1522
+ "eval_samples_per_second": 46.876,
1523
+ "eval_steps_per_second": 2.467,
1524
+ "step": 5050
1525
+ },
1526
+ {
1527
+ "epoch": 4.08,
1528
+ "grad_norm": 1.4417035579681396,
1529
+ "learning_rate": 2.213436313605413e-05,
1530
+ "loss": 0.1318,
1531
+ "step": 5100
1532
+ },
1533
+ {
1534
+ "epoch": 4.08,
1535
+ "eval_loss": 1.4657115936279297,
1536
+ "eval_runtime": 2.4283,
1537
+ "eval_samples_per_second": 46.947,
1538
+ "eval_steps_per_second": 2.471,
1539
+ "step": 5100
1540
+ },
1541
+ {
1542
+ "epoch": 4.12,
1543
+ "grad_norm": 1.5279566049575806,
1544
+ "learning_rate": 2.1949440526797928e-05,
1545
+ "loss": 0.1365,
1546
+ "step": 5150
1547
+ },
1548
+ {
1549
+ "epoch": 4.12,
1550
+ "eval_loss": 1.4270943403244019,
1551
+ "eval_runtime": 2.429,
1552
+ "eval_samples_per_second": 46.933,
1553
+ "eval_steps_per_second": 2.47,
1554
+ "step": 5150
1555
+ },
1556
+ {
1557
+ "epoch": 4.16,
1558
+ "grad_norm": 1.295238971710205,
1559
+ "learning_rate": 2.176316311034146e-05,
1560
+ "loss": 0.1289,
1561
+ "step": 5200
1562
+ },
1563
+ {
1564
+ "epoch": 4.16,
1565
+ "eval_loss": 1.4580848217010498,
1566
+ "eval_runtime": 2.4251,
1567
+ "eval_samples_per_second": 47.009,
1568
+ "eval_steps_per_second": 2.474,
1569
+ "step": 5200
1570
+ },
1571
+ {
1572
+ "epoch": 4.2,
1573
+ "grad_norm": 1.3161300420761108,
1574
+ "learning_rate": 2.157556720183616e-05,
1575
+ "loss": 0.1335,
1576
+ "step": 5250
1577
+ },
1578
+ {
1579
+ "epoch": 4.2,
1580
+ "eval_loss": 1.507900595664978,
1581
+ "eval_runtime": 2.4298,
1582
+ "eval_samples_per_second": 46.918,
1583
+ "eval_steps_per_second": 2.469,
1584
+ "step": 5250
1585
+ },
1586
+ {
1587
+ "epoch": 4.24,
1588
+ "grad_norm": 1.3103713989257812,
1589
+ "learning_rate": 2.138668937347609e-05,
1590
+ "loss": 0.1309,
1591
+ "step": 5300
1592
+ },
1593
+ {
1594
+ "epoch": 4.24,
1595
+ "eval_loss": 1.6110438108444214,
1596
+ "eval_runtime": 2.4254,
1597
+ "eval_samples_per_second": 47.002,
1598
+ "eval_steps_per_second": 2.474,
1599
+ "step": 5300
1600
+ },
1601
+ {
1602
+ "epoch": 4.28,
1603
+ "grad_norm": 1.3937710523605347,
1604
+ "learning_rate": 2.119656644736813e-05,
1605
+ "loss": 0.1424,
1606
+ "step": 5350
1607
+ },
1608
+ {
1609
+ "epoch": 4.28,
1610
+ "eval_loss": 1.6548389196395874,
1611
+ "eval_runtime": 2.4402,
1612
+ "eval_samples_per_second": 46.718,
1613
+ "eval_steps_per_second": 2.459,
1614
+ "step": 5350
1615
+ },
1616
+ {
1617
+ "epoch": 4.32,
1618
+ "grad_norm": 1.387892484664917,
1619
+ "learning_rate": 2.100523548835343e-05,
1620
+ "loss": 0.1373,
1621
+ "step": 5400
1622
+ },
1623
+ {
1624
+ "epoch": 4.32,
1625
+ "eval_loss": 1.5321193933486938,
1626
+ "eval_runtime": 2.4338,
1627
+ "eval_samples_per_second": 46.841,
1628
+ "eval_steps_per_second": 2.465,
1629
+ "step": 5400
1630
+ },
1631
+ {
1632
+ "epoch": 4.36,
1633
+ "grad_norm": 1.3102384805679321,
1634
+ "learning_rate": 2.0812733796781544e-05,
1635
+ "loss": 0.1345,
1636
+ "step": 5450
1637
+ },
1638
+ {
1639
+ "epoch": 4.36,
1640
+ "eval_loss": 1.538265585899353,
1641
+ "eval_runtime": 2.4296,
1642
+ "eval_samples_per_second": 46.921,
1643
+ "eval_steps_per_second": 2.47,
1644
+ "step": 5450
1645
+ },
1646
+ {
1647
+ "epoch": 4.4,
1648
+ "grad_norm": 1.616732120513916,
1649
+ "learning_rate": 2.0619098901238684e-05,
1650
+ "loss": 0.1305,
1651
+ "step": 5500
1652
+ },
1653
+ {
1654
+ "epoch": 4.4,
1655
+ "eval_loss": 1.50411856174469,
1656
+ "eval_runtime": 2.4345,
1657
+ "eval_samples_per_second": 46.828,
1658
+ "eval_steps_per_second": 2.465,
1659
+ "step": 5500
1660
+ },
1661
+ {
1662
+ "epoch": 4.44,
1663
+ "grad_norm": 1.4710546731948853,
1664
+ "learning_rate": 2.0424368551231383e-05,
1665
+ "loss": 0.1332,
1666
+ "step": 5550
1667
+ },
1668
+ {
1669
+ "epoch": 4.44,
1670
+ "eval_loss": 1.5596922636032104,
1671
+ "eval_runtime": 2.4348,
1672
+ "eval_samples_per_second": 46.821,
1673
+ "eval_steps_per_second": 2.464,
1674
+ "step": 5550
1675
+ },
1676
+ {
1677
+ "epoch": 4.48,
1678
+ "grad_norm": 1.2584961652755737,
1679
+ "learning_rate": 2.022858070982723e-05,
1680
+ "loss": 0.1394,
1681
+ "step": 5600
1682
+ },
1683
+ {
1684
+ "epoch": 4.48,
1685
+ "eval_loss": 1.553667664527893,
1686
+ "eval_runtime": 2.4307,
1687
+ "eval_samples_per_second": 46.901,
1688
+ "eval_steps_per_second": 2.468,
1689
+ "step": 5600
1690
+ },
1691
+ {
1692
+ "epoch": 4.52,
1693
+ "grad_norm": 1.3744111061096191,
1694
+ "learning_rate": 2.0031773546253828e-05,
1695
+ "loss": 0.1342,
1696
+ "step": 5650
1697
+ },
1698
+ {
1699
+ "epoch": 4.52,
1700
+ "eval_loss": 1.5672773122787476,
1701
+ "eval_runtime": 2.4307,
1702
+ "eval_samples_per_second": 46.9,
1703
+ "eval_steps_per_second": 2.468,
1704
+ "step": 5650
1705
+ },
1706
+ {
1707
+ "epoch": 4.5600000000000005,
1708
+ "grad_norm": 1.5977177619934082,
1709
+ "learning_rate": 1.983398542845767e-05,
1710
+ "loss": 0.142,
1711
+ "step": 5700
1712
+ },
1713
+ {
1714
+ "epoch": 4.5600000000000005,
1715
+ "eval_loss": 1.4725748300552368,
1716
+ "eval_runtime": 2.4372,
1717
+ "eval_samples_per_second": 46.774,
1718
+ "eval_steps_per_second": 2.462,
1719
+ "step": 5700
1720
+ },
1721
+ {
1722
+ "epoch": 4.6,
1723
+ "grad_norm": 1.4369518756866455,
1724
+ "learning_rate": 1.963525491562421e-05,
1725
+ "loss": 0.1386,
1726
+ "step": 5750
1727
+ },
1728
+ {
1729
+ "epoch": 4.6,
1730
+ "eval_loss": 1.4581120014190674,
1731
+ "eval_runtime": 2.4261,
1732
+ "eval_samples_per_second": 46.989,
1733
+ "eval_steps_per_second": 2.473,
1734
+ "step": 5750
1735
+ },
1736
+ {
1737
+ "epoch": 4.64,
1738
+ "grad_norm": 2.1108593940734863,
1739
+ "learning_rate": 1.9435620750660702e-05,
1740
+ "loss": 0.1384,
1741
+ "step": 5800
1742
+ },
1743
+ {
1744
+ "epoch": 4.64,
1745
+ "eval_loss": 1.3817390203475952,
1746
+ "eval_runtime": 2.4271,
1747
+ "eval_samples_per_second": 46.97,
1748
+ "eval_steps_per_second": 2.472,
1749
+ "step": 5800
1750
+ },
1751
+ {
1752
+ "epoch": 4.68,
1753
+ "grad_norm": 1.4125419855117798,
1754
+ "learning_rate": 1.923512185264315e-05,
1755
+ "loss": 0.1387,
1756
+ "step": 5850
1757
+ },
1758
+ {
1759
+ "epoch": 4.68,
1760
+ "eval_loss": 1.383178949356079,
1761
+ "eval_runtime": 2.4259,
1762
+ "eval_samples_per_second": 46.992,
1763
+ "eval_steps_per_second": 2.473,
1764
+ "step": 5850
1765
+ },
1766
+ {
1767
+ "epoch": 4.72,
1768
+ "grad_norm": 1.3931101560592651,
1769
+ "learning_rate": 1.9033797309228984e-05,
1770
+ "loss": 0.1391,
1771
+ "step": 5900
1772
+ },
1773
+ {
1774
+ "epoch": 4.72,
1775
+ "eval_loss": 1.393915057182312,
1776
+ "eval_runtime": 2.4319,
1777
+ "eval_samples_per_second": 46.876,
1778
+ "eval_steps_per_second": 2.467,
1779
+ "step": 5900
1780
+ },
1781
+ {
1782
+ "epoch": 4.76,
1783
+ "grad_norm": 1.8997610807418823,
1784
+ "learning_rate": 1.883168636903686e-05,
1785
+ "loss": 0.1428,
1786
+ "step": 5950
1787
+ },
1788
+ {
1789
+ "epoch": 4.76,
1790
+ "eval_loss": 1.3878700733184814,
1791
+ "eval_runtime": 2.4323,
1792
+ "eval_samples_per_second": 46.869,
1793
+ "eval_steps_per_second": 2.467,
1794
+ "step": 5950
1795
+ },
1796
+ {
1797
+ "epoch": 4.8,
1798
+ "grad_norm": 1.544852375984192,
1799
+ "learning_rate": 1.8628828433995013e-05,
1800
+ "loss": 0.1377,
1801
+ "step": 6000
1802
+ },
1803
+ {
1804
+ "epoch": 4.8,
1805
+ "eval_loss": 1.4498850107192993,
1806
+ "eval_runtime": 2.4376,
1807
+ "eval_samples_per_second": 46.767,
1808
+ "eval_steps_per_second": 2.461,
1809
+ "step": 6000
1810
  }
1811
  ],
1812
  "logging_steps": 50,
 
1814
  "num_input_tokens_seen": 0,
1815
  "num_train_epochs": 10,
1816
  "save_steps": 2000,
1817
+ "total_flos": 3.86553117644161e+17,
1818
  "train_batch_size": 4,
1819
  "trial_name": null,
1820
  "trial_params": null