| { | |
| "best_metric": 0.9585338620875524, | |
| "best_model_checkpoint": "vit-msn-small-wbc-classifier-mono-V-all/checkpoint-1040", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1040, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04807692307692308, | |
| "grad_norm": 4.224180221557617, | |
| "learning_rate": 4.807692307692308e-06, | |
| "loss": 0.5899, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09615384615384616, | |
| "grad_norm": 1.716335654258728, | |
| "learning_rate": 9.615384615384616e-06, | |
| "loss": 0.2795, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14423076923076922, | |
| "grad_norm": 1.9335592985153198, | |
| "learning_rate": 1.4423076923076923e-05, | |
| "loss": 0.2528, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19230769230769232, | |
| "grad_norm": 16.045499801635742, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 0.248, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2403846153846154, | |
| "grad_norm": 3.4744293689727783, | |
| "learning_rate": 2.4038461538461542e-05, | |
| "loss": 0.2172, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.28846153846153844, | |
| "grad_norm": 7.248939514160156, | |
| "learning_rate": 2.8846153846153845e-05, | |
| "loss": 0.1973, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.33653846153846156, | |
| "grad_norm": 13.428326606750488, | |
| "learning_rate": 3.365384615384616e-05, | |
| "loss": 0.245, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 14.730627059936523, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.23, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4326923076923077, | |
| "grad_norm": 6.623648643493652, | |
| "learning_rate": 4.326923076923077e-05, | |
| "loss": 0.1977, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4807692307692308, | |
| "grad_norm": 3.5982465744018555, | |
| "learning_rate": 4.8076923076923084e-05, | |
| "loss": 0.1912, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5288461538461539, | |
| "grad_norm": 8.891210556030273, | |
| "learning_rate": 4.9679487179487185e-05, | |
| "loss": 0.1757, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5769230769230769, | |
| "grad_norm": 16.044160842895508, | |
| "learning_rate": 4.9145299145299147e-05, | |
| "loss": 0.2119, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 4.071866989135742, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.1954, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6730769230769231, | |
| "grad_norm": 5.397825241088867, | |
| "learning_rate": 4.8076923076923084e-05, | |
| "loss": 0.1927, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7211538461538461, | |
| "grad_norm": 2.4581189155578613, | |
| "learning_rate": 4.7542735042735045e-05, | |
| "loss": 0.1921, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 4.148955345153809, | |
| "learning_rate": 4.700854700854701e-05, | |
| "loss": 0.1945, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.8173076923076923, | |
| "grad_norm": 7.619889259338379, | |
| "learning_rate": 4.6474358974358976e-05, | |
| "loss": 0.1904, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8653846153846154, | |
| "grad_norm": 8.802661895751953, | |
| "learning_rate": 4.594017094017094e-05, | |
| "loss": 0.1801, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.9134615384615384, | |
| "grad_norm": 1.7985179424285889, | |
| "learning_rate": 4.5405982905982906e-05, | |
| "loss": 0.1564, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 7.868402481079102, | |
| "learning_rate": 4.4871794871794874e-05, | |
| "loss": 0.1974, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9371655180379626, | |
| "eval_loss": 0.16416136920452118, | |
| "eval_runtime": 85.3977, | |
| "eval_samples_per_second": 304.142, | |
| "eval_steps_per_second": 4.754, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 1.0096153846153846, | |
| "grad_norm": 4.705746650695801, | |
| "learning_rate": 4.4337606837606836e-05, | |
| "loss": 0.203, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.0576923076923077, | |
| "grad_norm": 3.6939737796783447, | |
| "learning_rate": 4.3803418803418805e-05, | |
| "loss": 0.1752, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.1057692307692308, | |
| "grad_norm": 3.589677333831787, | |
| "learning_rate": 4.326923076923077e-05, | |
| "loss": 0.1858, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 6.519199848175049, | |
| "learning_rate": 4.2735042735042735e-05, | |
| "loss": 0.1814, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.2019230769230769, | |
| "grad_norm": 8.503952026367188, | |
| "learning_rate": 4.2200854700854704e-05, | |
| "loss": 0.1783, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.7003586292266846, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.1611, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.2980769230769231, | |
| "grad_norm": 2.9637396335601807, | |
| "learning_rate": 4.1132478632478634e-05, | |
| "loss": 0.1807, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.3461538461538463, | |
| "grad_norm": 2.4124629497528076, | |
| "learning_rate": 4.05982905982906e-05, | |
| "loss": 0.1669, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.3942307692307692, | |
| "grad_norm": 5.0713791847229, | |
| "learning_rate": 4.006410256410257e-05, | |
| "loss": 0.1743, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.4423076923076923, | |
| "grad_norm": 2.7951672077178955, | |
| "learning_rate": 3.952991452991453e-05, | |
| "loss": 0.1717, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.4903846153846154, | |
| "grad_norm": 4.324550628662109, | |
| "learning_rate": 3.89957264957265e-05, | |
| "loss": 0.188, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 2.242941379547119, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.1914, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.5865384615384617, | |
| "grad_norm": 2.958832025527954, | |
| "learning_rate": 3.7927350427350425e-05, | |
| "loss": 0.1527, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.6346153846153846, | |
| "grad_norm": 3.1927058696746826, | |
| "learning_rate": 3.739316239316239e-05, | |
| "loss": 0.1822, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.6826923076923077, | |
| "grad_norm": 5.284717082977295, | |
| "learning_rate": 3.685897435897436e-05, | |
| "loss": 0.1587, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.7307692307692308, | |
| "grad_norm": 2.2974815368652344, | |
| "learning_rate": 3.6324786324786323e-05, | |
| "loss": 0.1494, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.7788461538461537, | |
| "grad_norm": 2.8491830825805664, | |
| "learning_rate": 3.579059829059829e-05, | |
| "loss": 0.1556, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.8269230769230769, | |
| "grad_norm": 2.7790327072143555, | |
| "learning_rate": 3.525641025641026e-05, | |
| "loss": 0.1552, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 2.0398130416870117, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.1704, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 2.2376208305358887, | |
| "learning_rate": 3.418803418803419e-05, | |
| "loss": 0.1697, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.9711538461538463, | |
| "grad_norm": 2.9672436714172363, | |
| "learning_rate": 3.365384615384616e-05, | |
| "loss": 0.1589, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.950833557925538, | |
| "eval_loss": 0.1333734393119812, | |
| "eval_runtime": 85.0927, | |
| "eval_samples_per_second": 305.232, | |
| "eval_steps_per_second": 4.771, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 2.019230769230769, | |
| "grad_norm": 1.8962812423706055, | |
| "learning_rate": 3.311965811965812e-05, | |
| "loss": 0.1603, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.0673076923076925, | |
| "grad_norm": 3.2972991466522217, | |
| "learning_rate": 3.258547008547009e-05, | |
| "loss": 0.1524, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.1153846153846154, | |
| "grad_norm": 2.4130704402923584, | |
| "learning_rate": 3.205128205128206e-05, | |
| "loss": 0.1507, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.1634615384615383, | |
| "grad_norm": 4.791253089904785, | |
| "learning_rate": 3.151709401709402e-05, | |
| "loss": 0.1717, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.2115384615384617, | |
| "grad_norm": 3.113145112991333, | |
| "learning_rate": 3.098290598290599e-05, | |
| "loss": 0.1484, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.2596153846153846, | |
| "grad_norm": 2.3643486499786377, | |
| "learning_rate": 3.0448717948717947e-05, | |
| "loss": 0.1717, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 3.9917261600494385, | |
| "learning_rate": 2.9914529914529915e-05, | |
| "loss": 0.1647, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.355769230769231, | |
| "grad_norm": 7.72503662109375, | |
| "learning_rate": 2.9380341880341884e-05, | |
| "loss": 0.1506, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.4038461538461537, | |
| "grad_norm": 2.822436809539795, | |
| "learning_rate": 2.8846153846153845e-05, | |
| "loss": 0.161, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.451923076923077, | |
| "grad_norm": 2.8079020977020264, | |
| "learning_rate": 2.8311965811965814e-05, | |
| "loss": 0.132, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.609222650527954, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1531, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.5480769230769234, | |
| "grad_norm": 4.114030361175537, | |
| "learning_rate": 2.724358974358974e-05, | |
| "loss": 0.137, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.5961538461538463, | |
| "grad_norm": 2.093318223953247, | |
| "learning_rate": 2.670940170940171e-05, | |
| "loss": 0.1394, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.644230769230769, | |
| "grad_norm": 2.208742618560791, | |
| "learning_rate": 2.6175213675213678e-05, | |
| "loss": 0.1536, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 2.9571166038513184, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 0.1576, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.7403846153846154, | |
| "grad_norm": 2.523430824279785, | |
| "learning_rate": 2.5106837606837608e-05, | |
| "loss": 0.1486, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.7884615384615383, | |
| "grad_norm": 3.496368885040283, | |
| "learning_rate": 2.4572649572649573e-05, | |
| "loss": 0.1463, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.8365384615384617, | |
| "grad_norm": 2.219247579574585, | |
| "learning_rate": 2.4038461538461542e-05, | |
| "loss": 0.1464, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 2.0809261798858643, | |
| "learning_rate": 2.3504273504273504e-05, | |
| "loss": 0.1467, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9326923076923075, | |
| "grad_norm": 2.1210248470306396, | |
| "learning_rate": 2.297008547008547e-05, | |
| "loss": 0.1557, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.980769230769231, | |
| "grad_norm": 1.880336046218872, | |
| "learning_rate": 2.2435897435897437e-05, | |
| "loss": 0.134, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9431332537635236, | |
| "eval_loss": 0.14658282697200775, | |
| "eval_runtime": 85.5913, | |
| "eval_samples_per_second": 303.454, | |
| "eval_steps_per_second": 4.743, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 3.0288461538461537, | |
| "grad_norm": 1.8079543113708496, | |
| "learning_rate": 2.1901709401709402e-05, | |
| "loss": 0.1553, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.4903086423873901, | |
| "learning_rate": 2.1367521367521368e-05, | |
| "loss": 0.1485, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 2.690070629119873, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.1423, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.173076923076923, | |
| "grad_norm": 5.739427089691162, | |
| "learning_rate": 2.02991452991453e-05, | |
| "loss": 0.1471, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.2211538461538463, | |
| "grad_norm": 2.2277045249938965, | |
| "learning_rate": 1.9764957264957266e-05, | |
| "loss": 0.1476, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.269230769230769, | |
| "grad_norm": 1.629521369934082, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 0.1465, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.3173076923076925, | |
| "grad_norm": 1.9700013399124146, | |
| "learning_rate": 1.8696581196581197e-05, | |
| "loss": 0.1317, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 3.3653846153846154, | |
| "grad_norm": 4.258269309997559, | |
| "learning_rate": 1.8162393162393162e-05, | |
| "loss": 0.1329, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.4134615384615383, | |
| "grad_norm": 6.2838664054870605, | |
| "learning_rate": 1.762820512820513e-05, | |
| "loss": 0.1686, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 4.847503185272217, | |
| "learning_rate": 1.7094017094017095e-05, | |
| "loss": 0.1422, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.5096153846153846, | |
| "grad_norm": 2.352872371673584, | |
| "learning_rate": 1.655982905982906e-05, | |
| "loss": 0.1517, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.5576923076923075, | |
| "grad_norm": 2.504417896270752, | |
| "learning_rate": 1.602564102564103e-05, | |
| "loss": 0.1391, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 3.605769230769231, | |
| "grad_norm": 2.708151340484619, | |
| "learning_rate": 1.5491452991452994e-05, | |
| "loss": 0.118, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.6538461538461537, | |
| "grad_norm": 2.629040002822876, | |
| "learning_rate": 1.4957264957264958e-05, | |
| "loss": 0.146, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 3.7019230769230766, | |
| "grad_norm": 3.0288608074188232, | |
| "learning_rate": 1.4423076923076923e-05, | |
| "loss": 0.1261, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 2.5658912658691406, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.1533, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 3.7980769230769234, | |
| "grad_norm": 2.7179951667785645, | |
| "learning_rate": 1.3354700854700855e-05, | |
| "loss": 0.1413, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.7115960121154785, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 0.1417, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.894230769230769, | |
| "grad_norm": 2.939439535140991, | |
| "learning_rate": 1.2286324786324787e-05, | |
| "loss": 0.1311, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.9423076923076925, | |
| "grad_norm": 3.047938585281372, | |
| "learning_rate": 1.1752136752136752e-05, | |
| "loss": 0.1247, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.9903846153846154, | |
| "grad_norm": 2.0042459964752197, | |
| "learning_rate": 1.1217948717948719e-05, | |
| "loss": 0.1488, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9566472875678589, | |
| "eval_loss": 0.11553934961557388, | |
| "eval_runtime": 85.0472, | |
| "eval_samples_per_second": 305.395, | |
| "eval_steps_per_second": 4.774, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 4.038461538461538, | |
| "grad_norm": 2.6015453338623047, | |
| "learning_rate": 1.0683760683760684e-05, | |
| "loss": 0.1375, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.086538461538462, | |
| "grad_norm": 2.7020106315612793, | |
| "learning_rate": 1.014957264957265e-05, | |
| "loss": 0.1389, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.134615384615385, | |
| "grad_norm": 2.7265429496765137, | |
| "learning_rate": 9.615384615384616e-06, | |
| "loss": 0.1327, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 4.1826923076923075, | |
| "grad_norm": 5.0143866539001465, | |
| "learning_rate": 9.081196581196581e-06, | |
| "loss": 0.1436, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 3.4509687423706055, | |
| "learning_rate": 8.547008547008548e-06, | |
| "loss": 0.1238, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 4.278846153846154, | |
| "grad_norm": 2.778099536895752, | |
| "learning_rate": 8.012820512820515e-06, | |
| "loss": 0.1219, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 4.326923076923077, | |
| "grad_norm": 2.5686230659484863, | |
| "learning_rate": 7.478632478632479e-06, | |
| "loss": 0.1291, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 2.001837730407715, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.1234, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 4.423076923076923, | |
| "grad_norm": 2.5673305988311768, | |
| "learning_rate": 6.41025641025641e-06, | |
| "loss": 0.1279, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 4.471153846153846, | |
| "grad_norm": 3.475268602371216, | |
| "learning_rate": 5.876068376068376e-06, | |
| "loss": 0.1272, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 4.519230769230769, | |
| "grad_norm": 2.018739700317383, | |
| "learning_rate": 5.341880341880342e-06, | |
| "loss": 0.135, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 4.5673076923076925, | |
| "grad_norm": 3.1651690006256104, | |
| "learning_rate": 4.807692307692308e-06, | |
| "loss": 0.1429, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 1.8769733905792236, | |
| "learning_rate": 4.273504273504274e-06, | |
| "loss": 0.1254, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 4.663461538461538, | |
| "grad_norm": 2.2759199142456055, | |
| "learning_rate": 3.7393162393162394e-06, | |
| "loss": 0.1255, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 4.711538461538462, | |
| "grad_norm": 2.8174450397491455, | |
| "learning_rate": 3.205128205128205e-06, | |
| "loss": 0.1325, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 4.759615384615385, | |
| "grad_norm": 3.362974166870117, | |
| "learning_rate": 2.670940170940171e-06, | |
| "loss": 0.1334, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 4.8076923076923075, | |
| "grad_norm": 2.34413743019104, | |
| "learning_rate": 2.136752136752137e-06, | |
| "loss": 0.125, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.855769230769231, | |
| "grad_norm": 3.007368326187134, | |
| "learning_rate": 1.6025641025641025e-06, | |
| "loss": 0.1288, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 4.903846153846154, | |
| "grad_norm": 3.8937089443206787, | |
| "learning_rate": 1.0683760683760685e-06, | |
| "loss": 0.1272, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 4.951923076923077, | |
| "grad_norm": 2.394737958908081, | |
| "learning_rate": 5.341880341880342e-07, | |
| "loss": 0.1322, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.8422698974609375, | |
| "learning_rate": 0.0, | |
| "loss": 0.1169, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9585338620875524, | |
| "eval_loss": 0.11395391821861267, | |
| "eval_runtime": 84.4488, | |
| "eval_samples_per_second": 307.559, | |
| "eval_steps_per_second": 4.808, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 1040, | |
| "total_flos": 5.2079419237154e+18, | |
| "train_loss": 0.16386900280530636, | |
| "train_runtime": 2488.4676, | |
| "train_samples_per_second": 106.949, | |
| "train_steps_per_second": 0.418 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1040, | |
| "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": 5.2079419237154e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |