| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 100, | |
| "global_step": 2310, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.021645021645021644, | |
| "grad_norm": 2.8125383853912354, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9514, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04329004329004329, | |
| "grad_norm": 1.2800142765045166, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7706, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06493506493506493, | |
| "grad_norm": 1.0814083814620972, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6133, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08658008658008658, | |
| "grad_norm": 0.7970395088195801, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5192, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10822510822510822, | |
| "grad_norm": 0.6642207503318787, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4351, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12987012987012986, | |
| "grad_norm": 0.5111215114593506, | |
| "learning_rate": 0.00012, | |
| "loss": 0.3603, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "grad_norm": 1.2057693004608154, | |
| "learning_rate": 0.00014, | |
| "loss": 0.3361, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17316017316017315, | |
| "grad_norm": 0.4057958424091339, | |
| "learning_rate": 0.00016, | |
| "loss": 0.3053, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19480519480519481, | |
| "grad_norm": 0.5448099374771118, | |
| "learning_rate": 0.00018, | |
| "loss": 0.3024, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.21645021645021645, | |
| "grad_norm": 0.6636802554130554, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2595, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.21645021645021645, | |
| "eval_loss": 0.1525065004825592, | |
| "eval_runtime": 43.7956, | |
| "eval_samples_per_second": 11.417, | |
| "eval_steps_per_second": 0.365, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23809523809523808, | |
| "grad_norm": 0.6046785116195679, | |
| "learning_rate": 0.0001990950226244344, | |
| "loss": 0.2407, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2597402597402597, | |
| "grad_norm": 0.5757384896278381, | |
| "learning_rate": 0.0001981900452488688, | |
| "loss": 0.2223, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2813852813852814, | |
| "grad_norm": 0.5616349577903748, | |
| "learning_rate": 0.00019728506787330318, | |
| "loss": 0.224, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.30303030303030304, | |
| "grad_norm": 0.5419019460678101, | |
| "learning_rate": 0.00019638009049773755, | |
| "loss": 0.2306, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3246753246753247, | |
| "grad_norm": 0.49299636483192444, | |
| "learning_rate": 0.00019547511312217194, | |
| "loss": 0.2214, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3463203463203463, | |
| "grad_norm": 0.6301876306533813, | |
| "learning_rate": 0.00019457013574660634, | |
| "loss": 0.2296, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.36796536796536794, | |
| "grad_norm": 1.009000539779663, | |
| "learning_rate": 0.00019366515837104074, | |
| "loss": 0.2277, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.38961038961038963, | |
| "grad_norm": 0.5606774091720581, | |
| "learning_rate": 0.00019276018099547514, | |
| "loss": 0.2235, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.41125541125541126, | |
| "grad_norm": 0.47087350487709045, | |
| "learning_rate": 0.0001918552036199095, | |
| "loss": 0.2175, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4329004329004329, | |
| "grad_norm": 0.566985011100769, | |
| "learning_rate": 0.0001909502262443439, | |
| "loss": 0.2124, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4329004329004329, | |
| "eval_loss": 0.13294899463653564, | |
| "eval_runtime": 23.5705, | |
| "eval_samples_per_second": 21.213, | |
| "eval_steps_per_second": 0.679, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45454545454545453, | |
| "grad_norm": 0.7860066294670105, | |
| "learning_rate": 0.00019004524886877827, | |
| "loss": 0.2164, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 0.47214949131011963, | |
| "learning_rate": 0.00018914027149321267, | |
| "loss": 0.2086, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.49783549783549785, | |
| "grad_norm": 0.5978089570999146, | |
| "learning_rate": 0.00018823529411764707, | |
| "loss": 0.2176, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5194805194805194, | |
| "grad_norm": 0.5042030811309814, | |
| "learning_rate": 0.00018733031674208147, | |
| "loss": 0.2003, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5411255411255411, | |
| "grad_norm": 0.5709463953971863, | |
| "learning_rate": 0.00018642533936651584, | |
| "loss": 0.1969, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5627705627705628, | |
| "grad_norm": 0.5876468420028687, | |
| "learning_rate": 0.00018552036199095024, | |
| "loss": 0.1928, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5844155844155844, | |
| "grad_norm": 0.5380046963691711, | |
| "learning_rate": 0.00018461538461538463, | |
| "loss": 0.1944, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.6060606060606061, | |
| "grad_norm": 0.3912442624568939, | |
| "learning_rate": 0.000183710407239819, | |
| "loss": 0.1999, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6277056277056277, | |
| "grad_norm": 0.6305837631225586, | |
| "learning_rate": 0.0001828054298642534, | |
| "loss": 0.1976, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6493506493506493, | |
| "grad_norm": 0.6630299091339111, | |
| "learning_rate": 0.0001819004524886878, | |
| "loss": 0.1881, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6493506493506493, | |
| "eval_loss": 0.11815983802080154, | |
| "eval_runtime": 23.4543, | |
| "eval_samples_per_second": 21.318, | |
| "eval_steps_per_second": 0.682, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.670995670995671, | |
| "grad_norm": 0.3871740996837616, | |
| "learning_rate": 0.00018099547511312217, | |
| "loss": 0.1918, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6926406926406926, | |
| "grad_norm": 0.39743664860725403, | |
| "learning_rate": 0.00018009049773755657, | |
| "loss": 0.1875, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 0.580439031124115, | |
| "learning_rate": 0.00017918552036199096, | |
| "loss": 0.1881, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7359307359307359, | |
| "grad_norm": 0.39449596405029297, | |
| "learning_rate": 0.00017828054298642536, | |
| "loss": 0.189, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7575757575757576, | |
| "grad_norm": 0.5583558678627014, | |
| "learning_rate": 0.00017737556561085973, | |
| "loss": 0.1912, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7792207792207793, | |
| "grad_norm": 0.5783946514129639, | |
| "learning_rate": 0.00017647058823529413, | |
| "loss": 0.1891, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.8008658008658008, | |
| "grad_norm": 0.5039686560630798, | |
| "learning_rate": 0.0001755656108597285, | |
| "loss": 0.1859, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8225108225108225, | |
| "grad_norm": 0.45908617973327637, | |
| "learning_rate": 0.0001746606334841629, | |
| "loss": 0.1868, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8441558441558441, | |
| "grad_norm": 0.4211718738079071, | |
| "learning_rate": 0.0001737556561085973, | |
| "loss": 0.1961, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8658008658008658, | |
| "grad_norm": 0.4308678209781647, | |
| "learning_rate": 0.0001728506787330317, | |
| "loss": 0.2057, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8658008658008658, | |
| "eval_loss": 0.12011528760194778, | |
| "eval_runtime": 21.7151, | |
| "eval_samples_per_second": 23.025, | |
| "eval_steps_per_second": 0.737, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8874458874458875, | |
| "grad_norm": 0.5541667342185974, | |
| "learning_rate": 0.0001719457013574661, | |
| "loss": 0.1946, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 0.782439112663269, | |
| "learning_rate": 0.00017104072398190046, | |
| "loss": 0.1746, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9307359307359307, | |
| "grad_norm": 0.4267190396785736, | |
| "learning_rate": 0.00017013574660633486, | |
| "loss": 0.1905, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 0.4720573127269745, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 0.1925, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.974025974025974, | |
| "grad_norm": 0.49390169978141785, | |
| "learning_rate": 0.00016832579185520363, | |
| "loss": 0.1967, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9956709956709957, | |
| "grad_norm": 0.4374224841594696, | |
| "learning_rate": 0.00016742081447963802, | |
| "loss": 0.1813, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.0173160173160174, | |
| "grad_norm": 0.4421156048774719, | |
| "learning_rate": 0.00016651583710407242, | |
| "loss": 0.1673, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.0389610389610389, | |
| "grad_norm": 0.35764458775520325, | |
| "learning_rate": 0.0001656108597285068, | |
| "loss": 0.1759, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.0606060606060606, | |
| "grad_norm": 0.36364632844924927, | |
| "learning_rate": 0.0001647058823529412, | |
| "loss": 0.1726, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.0822510822510822, | |
| "grad_norm": 0.37476664781570435, | |
| "learning_rate": 0.00016380090497737556, | |
| "loss": 0.1819, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0822510822510822, | |
| "eval_loss": 0.12120614945888519, | |
| "eval_runtime": 38.4695, | |
| "eval_samples_per_second": 12.997, | |
| "eval_steps_per_second": 0.416, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.103896103896104, | |
| "grad_norm": 0.4027290344238281, | |
| "learning_rate": 0.00016289592760180996, | |
| "loss": 0.1633, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.1255411255411256, | |
| "grad_norm": 0.3566421866416931, | |
| "learning_rate": 0.00016199095022624435, | |
| "loss": 0.1607, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.1471861471861473, | |
| "grad_norm": 0.4911403954029083, | |
| "learning_rate": 0.00016108597285067875, | |
| "loss": 0.1655, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.1688311688311688, | |
| "grad_norm": 0.3717828392982483, | |
| "learning_rate": 0.00016018099547511315, | |
| "loss": 0.1646, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.1904761904761905, | |
| "grad_norm": 0.4402911961078644, | |
| "learning_rate": 0.00015927601809954752, | |
| "loss": 0.1677, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.2121212121212122, | |
| "grad_norm": 0.4106089770793915, | |
| "learning_rate": 0.0001583710407239819, | |
| "loss": 0.1607, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.2337662337662338, | |
| "grad_norm": 0.4335547387599945, | |
| "learning_rate": 0.0001574660633484163, | |
| "loss": 0.1625, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.2554112554112553, | |
| "grad_norm": 0.3730684816837311, | |
| "learning_rate": 0.00015656108597285069, | |
| "loss": 0.1627, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.277056277056277, | |
| "grad_norm": 0.4024712145328522, | |
| "learning_rate": 0.00015565610859728508, | |
| "loss": 0.1575, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.2987012987012987, | |
| "grad_norm": 0.5081992745399475, | |
| "learning_rate": 0.00015475113122171948, | |
| "loss": 0.1654, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.2987012987012987, | |
| "eval_loss": 0.11448249965906143, | |
| "eval_runtime": 25.9879, | |
| "eval_samples_per_second": 19.24, | |
| "eval_steps_per_second": 0.616, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3203463203463204, | |
| "grad_norm": 0.5350440144538879, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 0.158, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.341991341991342, | |
| "grad_norm": 0.44623634219169617, | |
| "learning_rate": 0.00015294117647058822, | |
| "loss": 0.1628, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.4370821714401245, | |
| "learning_rate": 0.00015203619909502262, | |
| "loss": 0.1563, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.3852813852813852, | |
| "grad_norm": 0.35637155175209045, | |
| "learning_rate": 0.00015113122171945702, | |
| "loss": 0.1706, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.406926406926407, | |
| "grad_norm": 0.3456045687198639, | |
| "learning_rate": 0.00015022624434389141, | |
| "loss": 0.1695, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.48406872153282166, | |
| "learning_rate": 0.0001493212669683258, | |
| "loss": 0.1561, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.4502164502164503, | |
| "grad_norm": 0.45908018946647644, | |
| "learning_rate": 0.00014841628959276018, | |
| "loss": 0.1584, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.4718614718614718, | |
| "grad_norm": 0.4222131073474884, | |
| "learning_rate": 0.00014751131221719458, | |
| "loss": 0.1524, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.4935064935064934, | |
| "grad_norm": 0.40552523732185364, | |
| "learning_rate": 0.00014660633484162895, | |
| "loss": 0.1664, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.5151515151515151, | |
| "grad_norm": 0.441180020570755, | |
| "learning_rate": 0.00014570135746606335, | |
| "loss": 0.1732, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5151515151515151, | |
| "eval_loss": 0.10767688602209091, | |
| "eval_runtime": 25.2267, | |
| "eval_samples_per_second": 19.82, | |
| "eval_steps_per_second": 0.634, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5367965367965368, | |
| "grad_norm": 0.39050132036209106, | |
| "learning_rate": 0.00014479638009049775, | |
| "loss": 0.1544, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.5584415584415585, | |
| "grad_norm": 0.37475523352622986, | |
| "learning_rate": 0.00014389140271493214, | |
| "loss": 0.1534, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.5800865800865802, | |
| "grad_norm": 0.3851136267185211, | |
| "learning_rate": 0.00014298642533936651, | |
| "loss": 0.1623, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.601731601731602, | |
| "grad_norm": 0.37124335765838623, | |
| "learning_rate": 0.0001420814479638009, | |
| "loss": 0.152, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.6233766233766234, | |
| "grad_norm": 0.3956158459186554, | |
| "learning_rate": 0.0001411764705882353, | |
| "loss": 0.1619, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.645021645021645, | |
| "grad_norm": 0.34193211793899536, | |
| "learning_rate": 0.00014027149321266968, | |
| "loss": 0.158, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.5510150194168091, | |
| "learning_rate": 0.00013936651583710408, | |
| "loss": 0.158, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.6883116883116882, | |
| "grad_norm": 0.6107664108276367, | |
| "learning_rate": 0.00013846153846153847, | |
| "loss": 0.1487, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.70995670995671, | |
| "grad_norm": 0.4111076295375824, | |
| "learning_rate": 0.00013755656108597284, | |
| "loss": 0.1588, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.7316017316017316, | |
| "grad_norm": 0.5250778198242188, | |
| "learning_rate": 0.00013665158371040724, | |
| "loss": 0.1653, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7316017316017316, | |
| "eval_loss": 0.10399862378835678, | |
| "eval_runtime": 23.9812, | |
| "eval_samples_per_second": 20.85, | |
| "eval_steps_per_second": 0.667, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7532467532467533, | |
| "grad_norm": 0.4260198771953583, | |
| "learning_rate": 0.00013574660633484164, | |
| "loss": 0.1482, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.774891774891775, | |
| "grad_norm": 0.28968125581741333, | |
| "learning_rate": 0.00013484162895927604, | |
| "loss": 0.1541, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.7965367965367967, | |
| "grad_norm": 0.29932093620300293, | |
| "learning_rate": 0.0001339366515837104, | |
| "loss": 0.1625, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.4358128011226654, | |
| "learning_rate": 0.0001330316742081448, | |
| "loss": 0.1658, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.8398268398268398, | |
| "grad_norm": 0.4161946773529053, | |
| "learning_rate": 0.00013212669683257918, | |
| "loss": 0.1472, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.8614718614718615, | |
| "grad_norm": 0.3558347225189209, | |
| "learning_rate": 0.00013122171945701357, | |
| "loss": 0.1671, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.883116883116883, | |
| "grad_norm": 0.41748765110969543, | |
| "learning_rate": 0.00013031674208144797, | |
| "loss": 0.1664, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.5195390582084656, | |
| "learning_rate": 0.00012941176470588237, | |
| "loss": 0.1603, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.9264069264069263, | |
| "grad_norm": 0.344159334897995, | |
| "learning_rate": 0.00012850678733031677, | |
| "loss": 0.1589, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.948051948051948, | |
| "grad_norm": 0.4217064380645752, | |
| "learning_rate": 0.00012760180995475114, | |
| "loss": 0.1631, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.948051948051948, | |
| "eval_loss": 0.11024898290634155, | |
| "eval_runtime": 26.0192, | |
| "eval_samples_per_second": 19.217, | |
| "eval_steps_per_second": 0.615, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.9696969696969697, | |
| "grad_norm": 0.377990186214447, | |
| "learning_rate": 0.0001266968325791855, | |
| "loss": 0.162, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.9913419913419914, | |
| "grad_norm": 0.3964829742908478, | |
| "learning_rate": 0.0001257918552036199, | |
| "loss": 0.1565, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.012987012987013, | |
| "grad_norm": 0.31516608595848083, | |
| "learning_rate": 0.0001248868778280543, | |
| "loss": 0.1356, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.034632034632035, | |
| "grad_norm": 0.4001309275627136, | |
| "learning_rate": 0.0001239819004524887, | |
| "loss": 0.1398, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.0562770562770565, | |
| "grad_norm": 0.2704612612724304, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 0.139, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.0779220779220777, | |
| "grad_norm": 0.37967827916145325, | |
| "learning_rate": 0.0001221719457013575, | |
| "loss": 0.1474, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.0995670995670994, | |
| "grad_norm": 0.3611961603164673, | |
| "learning_rate": 0.00012126696832579185, | |
| "loss": 0.1394, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.121212121212121, | |
| "grad_norm": 0.38165083527565, | |
| "learning_rate": 0.00012036199095022625, | |
| "loss": 0.1509, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 0.44539740681648254, | |
| "learning_rate": 0.00011945701357466063, | |
| "loss": 0.1335, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "grad_norm": 0.2796030044555664, | |
| "learning_rate": 0.00011855203619909503, | |
| "loss": 0.1354, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "eval_loss": 0.11003410071134567, | |
| "eval_runtime": 23.006, | |
| "eval_samples_per_second": 21.733, | |
| "eval_steps_per_second": 0.695, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.186147186147186, | |
| "grad_norm": 0.39886558055877686, | |
| "learning_rate": 0.00011764705882352942, | |
| "loss": 0.1365, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.207792207792208, | |
| "grad_norm": 0.27149999141693115, | |
| "learning_rate": 0.00011674208144796381, | |
| "loss": 0.1303, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.2294372294372296, | |
| "grad_norm": 0.4485577940940857, | |
| "learning_rate": 0.00011583710407239821, | |
| "loss": 0.136, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.2510822510822512, | |
| "grad_norm": 0.5114185214042664, | |
| "learning_rate": 0.00011493212669683258, | |
| "loss": 0.129, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.2727272727272725, | |
| "grad_norm": 0.41526681184768677, | |
| "learning_rate": 0.00011402714932126696, | |
| "loss": 0.1352, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.2943722943722946, | |
| "grad_norm": 0.4560807943344116, | |
| "learning_rate": 0.00011312217194570136, | |
| "loss": 0.1473, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.316017316017316, | |
| "grad_norm": 0.4193613529205322, | |
| "learning_rate": 0.00011221719457013576, | |
| "loss": 0.1427, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.3376623376623376, | |
| "grad_norm": 0.5285654664039612, | |
| "learning_rate": 0.00011131221719457014, | |
| "loss": 0.1395, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.3593073593073592, | |
| "grad_norm": 0.3672701120376587, | |
| "learning_rate": 0.00011040723981900454, | |
| "loss": 0.1303, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.4193953573703766, | |
| "learning_rate": 0.00010950226244343893, | |
| "loss": 0.1353, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "eval_loss": 0.10145355015993118, | |
| "eval_runtime": 20.0537, | |
| "eval_samples_per_second": 24.933, | |
| "eval_steps_per_second": 0.798, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.4025974025974026, | |
| "grad_norm": 0.39170539379119873, | |
| "learning_rate": 0.0001085972850678733, | |
| "loss": 0.1344, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.4242424242424243, | |
| "grad_norm": 0.4148479104042053, | |
| "learning_rate": 0.0001076923076923077, | |
| "loss": 0.1463, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.445887445887446, | |
| "grad_norm": 0.3758697211742401, | |
| "learning_rate": 0.00010678733031674209, | |
| "loss": 0.1338, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.4675324675324677, | |
| "grad_norm": 0.3020533621311188, | |
| "learning_rate": 0.00010588235294117647, | |
| "loss": 0.1369, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.4891774891774894, | |
| "grad_norm": 0.3161577582359314, | |
| "learning_rate": 0.00010497737556561087, | |
| "loss": 0.1438, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.5108225108225106, | |
| "grad_norm": 0.39273881912231445, | |
| "learning_rate": 0.00010407239819004526, | |
| "loss": 0.1404, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.5324675324675323, | |
| "grad_norm": 0.3449176251888275, | |
| "learning_rate": 0.00010316742081447965, | |
| "loss": 0.1346, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.554112554112554, | |
| "grad_norm": 0.261294960975647, | |
| "learning_rate": 0.00010226244343891402, | |
| "loss": 0.1244, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.5757575757575757, | |
| "grad_norm": 0.5024470686912537, | |
| "learning_rate": 0.00010135746606334842, | |
| "loss": 0.1385, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.5974025974025974, | |
| "grad_norm": 0.4657529294490814, | |
| "learning_rate": 0.0001004524886877828, | |
| "loss": 0.1458, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.5974025974025974, | |
| "eval_loss": 0.09968920797109604, | |
| "eval_runtime": 22.2909, | |
| "eval_samples_per_second": 22.431, | |
| "eval_steps_per_second": 0.718, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.619047619047619, | |
| "grad_norm": 0.28800147771835327, | |
| "learning_rate": 9.95475113122172e-05, | |
| "loss": 0.1366, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 2.6406926406926408, | |
| "grad_norm": 0.24663890898227692, | |
| "learning_rate": 9.864253393665159e-05, | |
| "loss": 0.1315, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.6623376623376624, | |
| "grad_norm": 0.3126956522464752, | |
| "learning_rate": 9.773755656108597e-05, | |
| "loss": 0.1305, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 2.683982683982684, | |
| "grad_norm": 0.34753990173339844, | |
| "learning_rate": 9.683257918552037e-05, | |
| "loss": 0.1326, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.7056277056277054, | |
| "grad_norm": 0.4045655429363251, | |
| "learning_rate": 9.592760180995475e-05, | |
| "loss": 0.147, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.369632750749588, | |
| "learning_rate": 9.502262443438914e-05, | |
| "loss": 0.1339, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.7489177489177488, | |
| "grad_norm": 0.3891206383705139, | |
| "learning_rate": 9.411764705882353e-05, | |
| "loss": 0.1326, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 2.7705627705627704, | |
| "grad_norm": 0.4095381498336792, | |
| "learning_rate": 9.321266968325792e-05, | |
| "loss": 0.1305, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.792207792207792, | |
| "grad_norm": 0.39312949776649475, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.1348, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.813852813852814, | |
| "grad_norm": 0.4769635498523712, | |
| "learning_rate": 9.14027149321267e-05, | |
| "loss": 0.1292, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.813852813852814, | |
| "eval_loss": 0.09773257374763489, | |
| "eval_runtime": 21.4232, | |
| "eval_samples_per_second": 23.339, | |
| "eval_steps_per_second": 0.747, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.8354978354978355, | |
| "grad_norm": 0.4419485628604889, | |
| "learning_rate": 9.049773755656108e-05, | |
| "loss": 0.1332, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.273708313703537, | |
| "learning_rate": 8.959276018099548e-05, | |
| "loss": 0.1351, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.878787878787879, | |
| "grad_norm": 0.41094347834587097, | |
| "learning_rate": 8.868778280542987e-05, | |
| "loss": 0.1436, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 2.9004329004329006, | |
| "grad_norm": 0.3220170736312866, | |
| "learning_rate": 8.778280542986425e-05, | |
| "loss": 0.1372, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.9220779220779223, | |
| "grad_norm": 0.3632793128490448, | |
| "learning_rate": 8.687782805429865e-05, | |
| "loss": 0.1421, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.9437229437229435, | |
| "grad_norm": 0.33830907940864563, | |
| "learning_rate": 8.597285067873304e-05, | |
| "loss": 0.1452, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.965367965367965, | |
| "grad_norm": 0.3863072097301483, | |
| "learning_rate": 8.506787330316743e-05, | |
| "loss": 0.1424, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.987012987012987, | |
| "grad_norm": 0.5766463279724121, | |
| "learning_rate": 8.416289592760181e-05, | |
| "loss": 0.1424, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 3.0086580086580086, | |
| "grad_norm": 0.23465487360954285, | |
| "learning_rate": 8.325791855203621e-05, | |
| "loss": 0.1241, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 3.0303030303030303, | |
| "grad_norm": 0.32335883378982544, | |
| "learning_rate": 8.23529411764706e-05, | |
| "loss": 0.1181, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.0303030303030303, | |
| "eval_loss": 0.10378438234329224, | |
| "eval_runtime": 24.145, | |
| "eval_samples_per_second": 20.708, | |
| "eval_steps_per_second": 0.663, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.051948051948052, | |
| "grad_norm": 0.3104758560657501, | |
| "learning_rate": 8.144796380090498e-05, | |
| "loss": 0.1227, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 3.0735930735930737, | |
| "grad_norm": 0.27100279927253723, | |
| "learning_rate": 8.054298642533938e-05, | |
| "loss": 0.1164, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 3.0952380952380953, | |
| "grad_norm": 0.3317118287086487, | |
| "learning_rate": 7.963800904977376e-05, | |
| "loss": 0.1233, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 3.116883116883117, | |
| "grad_norm": 0.41682133078575134, | |
| "learning_rate": 7.873303167420814e-05, | |
| "loss": 0.1213, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 3.1385281385281387, | |
| "grad_norm": 0.31365829706192017, | |
| "learning_rate": 7.782805429864254e-05, | |
| "loss": 0.1183, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.16017316017316, | |
| "grad_norm": 0.3687366247177124, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.119, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 3.1818181818181817, | |
| "grad_norm": 0.4626697599887848, | |
| "learning_rate": 7.601809954751131e-05, | |
| "loss": 0.1244, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 3.2034632034632033, | |
| "grad_norm": 0.46320992708206177, | |
| "learning_rate": 7.511312217194571e-05, | |
| "loss": 0.1214, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 3.225108225108225, | |
| "grad_norm": 0.3054867386817932, | |
| "learning_rate": 7.420814479638009e-05, | |
| "loss": 0.1276, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 3.2467532467532467, | |
| "grad_norm": 0.26975077390670776, | |
| "learning_rate": 7.330316742081448e-05, | |
| "loss": 0.1235, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.2467532467532467, | |
| "eval_loss": 0.10016042739152908, | |
| "eval_runtime": 23.6078, | |
| "eval_samples_per_second": 21.179, | |
| "eval_steps_per_second": 0.678, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.2683982683982684, | |
| "grad_norm": 0.33022555708885193, | |
| "learning_rate": 7.239819004524887e-05, | |
| "loss": 0.12, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 3.29004329004329, | |
| "grad_norm": 0.4828197956085205, | |
| "learning_rate": 7.149321266968326e-05, | |
| "loss": 0.1299, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 3.311688311688312, | |
| "grad_norm": 0.36965712904930115, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 0.1217, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.4338422417640686, | |
| "learning_rate": 6.968325791855204e-05, | |
| "loss": 0.1276, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 3.354978354978355, | |
| "grad_norm": 0.40615785121917725, | |
| "learning_rate": 6.877828054298642e-05, | |
| "loss": 0.1214, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.3766233766233764, | |
| "grad_norm": 0.2703055739402771, | |
| "learning_rate": 6.787330316742082e-05, | |
| "loss": 0.1179, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 3.398268398268398, | |
| "grad_norm": 0.469108521938324, | |
| "learning_rate": 6.69683257918552e-05, | |
| "loss": 0.1189, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 3.41991341991342, | |
| "grad_norm": 0.2688772976398468, | |
| "learning_rate": 6.606334841628959e-05, | |
| "loss": 0.1174, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 3.4415584415584415, | |
| "grad_norm": 0.2989065647125244, | |
| "learning_rate": 6.515837104072399e-05, | |
| "loss": 0.111, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 3.463203463203463, | |
| "grad_norm": 0.397502601146698, | |
| "learning_rate": 6.425339366515838e-05, | |
| "loss": 0.1189, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.463203463203463, | |
| "eval_loss": 0.10202713310718536, | |
| "eval_runtime": 25.0286, | |
| "eval_samples_per_second": 19.977, | |
| "eval_steps_per_second": 0.639, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.484848484848485, | |
| "grad_norm": 0.3283541202545166, | |
| "learning_rate": 6.334841628959275e-05, | |
| "loss": 0.1216, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 3.5064935064935066, | |
| "grad_norm": 0.3664211928844452, | |
| "learning_rate": 6.244343891402715e-05, | |
| "loss": 0.1171, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 3.5281385281385282, | |
| "grad_norm": 0.33489325642585754, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.1222, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 3.54978354978355, | |
| "grad_norm": 0.27272143959999084, | |
| "learning_rate": 6.0633484162895926e-05, | |
| "loss": 0.1089, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 0.3384559750556946, | |
| "learning_rate": 5.972850678733032e-05, | |
| "loss": 0.1092, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.5930735930735933, | |
| "grad_norm": 0.4882173240184784, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.1326, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 3.6147186147186146, | |
| "grad_norm": 0.33967503905296326, | |
| "learning_rate": 5.7918552036199105e-05, | |
| "loss": 0.119, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.30732062458992004, | |
| "learning_rate": 5.701357466063348e-05, | |
| "loss": 0.1115, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 3.658008658008658, | |
| "grad_norm": 0.46696531772613525, | |
| "learning_rate": 5.610859728506788e-05, | |
| "loss": 0.1246, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 3.6796536796536796, | |
| "grad_norm": 0.4071387052536011, | |
| "learning_rate": 5.520361990950227e-05, | |
| "loss": 0.1305, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.6796536796536796, | |
| "eval_loss": 0.09984961152076721, | |
| "eval_runtime": 22.2364, | |
| "eval_samples_per_second": 22.486, | |
| "eval_steps_per_second": 0.72, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.7012987012987013, | |
| "grad_norm": 0.34078463912010193, | |
| "learning_rate": 5.429864253393665e-05, | |
| "loss": 0.1227, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 3.722943722943723, | |
| "grad_norm": 0.46931177377700806, | |
| "learning_rate": 5.3393665158371045e-05, | |
| "loss": 0.1222, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 3.7445887445887447, | |
| "grad_norm": 0.28127625584602356, | |
| "learning_rate": 5.2488687782805436e-05, | |
| "loss": 0.1186, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 3.7662337662337664, | |
| "grad_norm": 0.3582036793231964, | |
| "learning_rate": 5.158371040723983e-05, | |
| "loss": 0.1124, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 3.787878787878788, | |
| "grad_norm": 0.31167203187942505, | |
| "learning_rate": 5.067873303167421e-05, | |
| "loss": 0.1179, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.391791969537735, | |
| "learning_rate": 4.97737556561086e-05, | |
| "loss": 0.127, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 3.8311688311688314, | |
| "grad_norm": 0.3683635890483856, | |
| "learning_rate": 4.8868778280542986e-05, | |
| "loss": 0.1129, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 3.8528138528138527, | |
| "grad_norm": 0.3955051302909851, | |
| "learning_rate": 4.7963800904977377e-05, | |
| "loss": 0.125, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 3.8744588744588744, | |
| "grad_norm": 0.40321120619773865, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 0.1272, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 3.896103896103896, | |
| "grad_norm": 0.31911927461624146, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.1072, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.896103896103896, | |
| "eval_loss": 0.10081314295530319, | |
| "eval_runtime": 21.9055, | |
| "eval_samples_per_second": 22.825, | |
| "eval_steps_per_second": 0.73, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.9177489177489178, | |
| "grad_norm": 0.4627375900745392, | |
| "learning_rate": 4.524886877828054e-05, | |
| "loss": 0.1238, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 3.9393939393939394, | |
| "grad_norm": 0.43323320150375366, | |
| "learning_rate": 4.434389140271493e-05, | |
| "loss": 0.1202, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 3.961038961038961, | |
| "grad_norm": 0.4214964807033539, | |
| "learning_rate": 4.3438914027149324e-05, | |
| "loss": 0.1186, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 3.982683982683983, | |
| "grad_norm": 0.302083283662796, | |
| "learning_rate": 4.2533936651583714e-05, | |
| "loss": 0.1226, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 4.004329004329004, | |
| "grad_norm": 0.39333033561706543, | |
| "learning_rate": 4.1628959276018105e-05, | |
| "loss": 0.1249, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 4.025974025974026, | |
| "grad_norm": 0.34783726930618286, | |
| "learning_rate": 4.072398190045249e-05, | |
| "loss": 0.1135, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 4.0476190476190474, | |
| "grad_norm": 0.3140253722667694, | |
| "learning_rate": 3.981900452488688e-05, | |
| "loss": 0.1095, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 4.06926406926407, | |
| "grad_norm": 0.3263933062553406, | |
| "learning_rate": 3.891402714932127e-05, | |
| "loss": 0.1136, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 4.090909090909091, | |
| "grad_norm": 0.4250706434249878, | |
| "learning_rate": 3.8009049773755655e-05, | |
| "loss": 0.1056, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 4.112554112554113, | |
| "grad_norm": 0.2250766158103943, | |
| "learning_rate": 3.7104072398190046e-05, | |
| "loss": 0.1079, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.112554112554113, | |
| "eval_loss": 0.10422435402870178, | |
| "eval_runtime": 22.3048, | |
| "eval_samples_per_second": 22.417, | |
| "eval_steps_per_second": 0.717, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.134199134199134, | |
| "grad_norm": 0.35859718918800354, | |
| "learning_rate": 3.6199095022624436e-05, | |
| "loss": 0.1048, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 4.1558441558441555, | |
| "grad_norm": 0.2876388728618622, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 0.1109, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 4.177489177489178, | |
| "grad_norm": 0.2952074408531189, | |
| "learning_rate": 3.438914027149321e-05, | |
| "loss": 0.1118, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 4.199134199134199, | |
| "grad_norm": 0.38420605659484863, | |
| "learning_rate": 3.34841628959276e-05, | |
| "loss": 0.1113, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 4.220779220779221, | |
| "grad_norm": 0.35064896941185, | |
| "learning_rate": 3.257918552036199e-05, | |
| "loss": 0.108, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 4.242424242424242, | |
| "grad_norm": 0.3417244553565979, | |
| "learning_rate": 3.167420814479638e-05, | |
| "loss": 0.1067, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 4.264069264069264, | |
| "grad_norm": 0.2506988048553467, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.116, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.40218839049339294, | |
| "learning_rate": 2.986425339366516e-05, | |
| "loss": 0.111, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 4.307359307359308, | |
| "grad_norm": 0.33777332305908203, | |
| "learning_rate": 2.8959276018099553e-05, | |
| "loss": 0.1081, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 4.329004329004329, | |
| "grad_norm": 0.2849682569503784, | |
| "learning_rate": 2.805429864253394e-05, | |
| "loss": 0.1043, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.329004329004329, | |
| "eval_loss": 0.10182041674852371, | |
| "eval_runtime": 32.5102, | |
| "eval_samples_per_second": 15.38, | |
| "eval_steps_per_second": 0.492, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.35064935064935, | |
| "grad_norm": 0.32567378878593445, | |
| "learning_rate": 2.7149321266968324e-05, | |
| "loss": 0.1065, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 4.372294372294372, | |
| "grad_norm": 0.35031840205192566, | |
| "learning_rate": 2.6244343891402718e-05, | |
| "loss": 0.1133, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 4.393939393939394, | |
| "grad_norm": 0.3233943581581116, | |
| "learning_rate": 2.5339366515837106e-05, | |
| "loss": 0.1034, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 4.415584415584416, | |
| "grad_norm": 0.326164573431015, | |
| "learning_rate": 2.4434389140271493e-05, | |
| "loss": 0.1053, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 4.437229437229437, | |
| "grad_norm": 0.3046077489852905, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 0.1034, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 4.458874458874459, | |
| "grad_norm": 0.3930221498012543, | |
| "learning_rate": 2.262443438914027e-05, | |
| "loss": 0.1132, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 4.48051948051948, | |
| "grad_norm": 0.4203428626060486, | |
| "learning_rate": 2.1719457013574662e-05, | |
| "loss": 0.1123, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 4.5021645021645025, | |
| "grad_norm": 0.23356233537197113, | |
| "learning_rate": 2.0814479638009053e-05, | |
| "loss": 0.1158, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 4.523809523809524, | |
| "grad_norm": 0.3921778202056885, | |
| "learning_rate": 1.990950226244344e-05, | |
| "loss": 0.1017, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 0.41227224469184875, | |
| "learning_rate": 1.9004524886877827e-05, | |
| "loss": 0.1095, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "eval_loss": 0.10190469026565552, | |
| "eval_runtime": 22.5368, | |
| "eval_samples_per_second": 22.186, | |
| "eval_steps_per_second": 0.71, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.567099567099567, | |
| "grad_norm": 0.46910232305526733, | |
| "learning_rate": 1.8099547511312218e-05, | |
| "loss": 0.1041, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 4.588744588744589, | |
| "grad_norm": 0.2650741934776306, | |
| "learning_rate": 1.7194570135746606e-05, | |
| "loss": 0.1096, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 4.6103896103896105, | |
| "grad_norm": 0.30446478724479675, | |
| "learning_rate": 1.6289592760180996e-05, | |
| "loss": 0.11, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 4.632034632034632, | |
| "grad_norm": 0.30164214968681335, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.1009, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 4.653679653679654, | |
| "grad_norm": 0.33579009771347046, | |
| "learning_rate": 1.4479638009049776e-05, | |
| "loss": 0.116, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 4.675324675324675, | |
| "grad_norm": 0.2805331349372864, | |
| "learning_rate": 1.3574660633484162e-05, | |
| "loss": 0.1154, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 4.696969696969697, | |
| "grad_norm": 0.2765107750892639, | |
| "learning_rate": 1.2669683257918553e-05, | |
| "loss": 0.1124, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 4.7186147186147185, | |
| "grad_norm": 0.39715567231178284, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 0.1078, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 4.740259740259741, | |
| "grad_norm": 0.4515060782432556, | |
| "learning_rate": 1.0859728506787331e-05, | |
| "loss": 0.1135, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.3021312654018402, | |
| "learning_rate": 9.95475113122172e-06, | |
| "loss": 0.1069, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "eval_loss": 0.10030877590179443, | |
| "eval_runtime": 22.0777, | |
| "eval_samples_per_second": 22.647, | |
| "eval_steps_per_second": 0.725, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.783549783549784, | |
| "grad_norm": 0.41326892375946045, | |
| "learning_rate": 9.049773755656109e-06, | |
| "loss": 0.1131, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 4.805194805194805, | |
| "grad_norm": 0.3018375039100647, | |
| "learning_rate": 8.144796380090498e-06, | |
| "loss": 0.1064, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 4.8268398268398265, | |
| "grad_norm": 0.3863151967525482, | |
| "learning_rate": 7.239819004524888e-06, | |
| "loss": 0.1187, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 4.848484848484849, | |
| "grad_norm": 0.47006744146347046, | |
| "learning_rate": 6.334841628959276e-06, | |
| "loss": 0.1073, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 4.87012987012987, | |
| "grad_norm": 0.36669015884399414, | |
| "learning_rate": 5.4298642533936655e-06, | |
| "loss": 0.1098, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 4.891774891774892, | |
| "grad_norm": 0.3454284965991974, | |
| "learning_rate": 4.5248868778280546e-06, | |
| "loss": 0.0987, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 4.913419913419913, | |
| "grad_norm": 0.45213282108306885, | |
| "learning_rate": 3.619909502262444e-06, | |
| "loss": 0.1094, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 4.935064935064935, | |
| "grad_norm": 0.3391339182853699, | |
| "learning_rate": 2.7149321266968327e-06, | |
| "loss": 0.1011, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 4.956709956709957, | |
| "grad_norm": 0.3675156235694885, | |
| "learning_rate": 1.809954751131222e-06, | |
| "loss": 0.1143, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 4.978354978354979, | |
| "grad_norm": 0.3254551589488983, | |
| "learning_rate": 9.04977375565611e-07, | |
| "loss": 0.1123, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.978354978354979, | |
| "eval_loss": 0.09990464895963669, | |
| "eval_runtime": 21.2785, | |
| "eval_samples_per_second": 23.498, | |
| "eval_steps_per_second": 0.752, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.34439313411712646, | |
| "learning_rate": 0.0, | |
| "loss": 0.1115, | |
| "step": 2310 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2310, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.35957960819826e+18, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |