| { | |
| "best_metric": 0.15439653396606445, | |
| "best_model_checkpoint": "./vit-base-brain-alzheimer-detection/checkpoint-4000", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 5120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1953125, | |
| "grad_norm": 3.019329786300659, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.3778, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.390625, | |
| "grad_norm": 2.534724712371826, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.1895, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5859375, | |
| "grad_norm": 2.774301052093506, | |
| "learning_rate": 6e-06, | |
| "loss": 1.0423, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "grad_norm": 3.057741165161133, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9859, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9765625, | |
| "grad_norm": 2.8204987049102783, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9008, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.171875, | |
| "grad_norm": 1.7199639081954956, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.8892, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3671875, | |
| "grad_norm": 3.138557195663452, | |
| "learning_rate": 1.4e-05, | |
| "loss": 0.8273, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 3.1744067668914795, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.7807, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.7578125, | |
| "grad_norm": 5.47639799118042, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.7546, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.953125, | |
| "grad_norm": 2.743036985397339, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7278, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.953125, | |
| "eval_accuracy": 0.705078125, | |
| "eval_loss": 0.7124308347702026, | |
| "eval_runtime": 5.7619, | |
| "eval_samples_per_second": 177.72, | |
| "eval_steps_per_second": 22.215, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.1484375, | |
| "grad_norm": 12.242274284362793, | |
| "learning_rate": 1.9783549783549785e-05, | |
| "loss": 0.6108, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.34375, | |
| "grad_norm": 9.961933135986328, | |
| "learning_rate": 1.9567099567099568e-05, | |
| "loss": 0.5815, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.5390625, | |
| "grad_norm": 8.35198974609375, | |
| "learning_rate": 1.9350649350649354e-05, | |
| "loss": 0.5702, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.734375, | |
| "grad_norm": 4.211214542388916, | |
| "learning_rate": 1.9134199134199138e-05, | |
| "loss": 0.4907, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.9296875, | |
| "grad_norm": 5.052712917327881, | |
| "learning_rate": 1.891774891774892e-05, | |
| "loss": 0.4993, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 2.4637234210968018, | |
| "learning_rate": 1.8701298701298704e-05, | |
| "loss": 0.3788, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.3203125, | |
| "grad_norm": 5.4455790519714355, | |
| "learning_rate": 1.8484848484848487e-05, | |
| "loss": 0.358, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.515625, | |
| "grad_norm": 13.753146171569824, | |
| "learning_rate": 1.826839826839827e-05, | |
| "loss": 0.273, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.7109375, | |
| "grad_norm": 7.315534591674805, | |
| "learning_rate": 1.8051948051948053e-05, | |
| "loss": 0.2588, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.90625, | |
| "grad_norm": 15.05672836303711, | |
| "learning_rate": 1.7835497835497836e-05, | |
| "loss": 0.3023, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.90625, | |
| "eval_accuracy": 0.8828125, | |
| "eval_loss": 0.3776271939277649, | |
| "eval_runtime": 6.9418, | |
| "eval_samples_per_second": 147.512, | |
| "eval_steps_per_second": 18.439, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.1015625, | |
| "grad_norm": 0.9234067797660828, | |
| "learning_rate": 1.761904761904762e-05, | |
| "loss": 0.2126, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 4.296875, | |
| "grad_norm": 4.910510540008545, | |
| "learning_rate": 1.7402597402597403e-05, | |
| "loss": 0.1539, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.4921875, | |
| "grad_norm": 6.401501655578613, | |
| "learning_rate": 1.718614718614719e-05, | |
| "loss": 0.2101, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 4.6875, | |
| "grad_norm": 1.9736899137496948, | |
| "learning_rate": 1.6969696969696972e-05, | |
| "loss": 0.1984, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.8828125, | |
| "grad_norm": 8.743949890136719, | |
| "learning_rate": 1.6753246753246756e-05, | |
| "loss": 0.1249, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.078125, | |
| "grad_norm": 14.517168998718262, | |
| "learning_rate": 1.653679653679654e-05, | |
| "loss": 0.1279, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 5.2734375, | |
| "grad_norm": 0.22373303771018982, | |
| "learning_rate": 1.6320346320346322e-05, | |
| "loss": 0.0986, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 5.46875, | |
| "grad_norm": 0.18486973643302917, | |
| "learning_rate": 1.6103896103896105e-05, | |
| "loss": 0.0922, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.6640625, | |
| "grad_norm": 0.4893491268157959, | |
| "learning_rate": 1.5887445887445888e-05, | |
| "loss": 0.0666, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 5.859375, | |
| "grad_norm": 17.192604064941406, | |
| "learning_rate": 1.567099567099567e-05, | |
| "loss": 0.0997, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.859375, | |
| "eval_accuracy": 0.9130859375, | |
| "eval_loss": 0.2808457016944885, | |
| "eval_runtime": 5.8114, | |
| "eval_samples_per_second": 176.206, | |
| "eval_steps_per_second": 22.026, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0546875, | |
| "grad_norm": 0.13603505492210388, | |
| "learning_rate": 1.5454545454545454e-05, | |
| "loss": 0.0481, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 44.868900299072266, | |
| "learning_rate": 1.523809523809524e-05, | |
| "loss": 0.0465, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.4453125, | |
| "grad_norm": 3.908478021621704, | |
| "learning_rate": 1.5021645021645024e-05, | |
| "loss": 0.0485, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 6.640625, | |
| "grad_norm": 0.24283438920974731, | |
| "learning_rate": 1.4805194805194807e-05, | |
| "loss": 0.0387, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 6.8359375, | |
| "grad_norm": 21.719118118286133, | |
| "learning_rate": 1.458874458874459e-05, | |
| "loss": 0.0803, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 7.03125, | |
| "grad_norm": 0.08735662698745728, | |
| "learning_rate": 1.4372294372294374e-05, | |
| "loss": 0.0266, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.2265625, | |
| "grad_norm": 0.16133151948451996, | |
| "learning_rate": 1.4155844155844157e-05, | |
| "loss": 0.0252, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 7.421875, | |
| "grad_norm": 0.12098684161901474, | |
| "learning_rate": 1.3939393939393942e-05, | |
| "loss": 0.0174, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 7.6171875, | |
| "grad_norm": 0.14388597011566162, | |
| "learning_rate": 1.3722943722943725e-05, | |
| "loss": 0.016, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 7.8125, | |
| "grad_norm": 0.07232346385717392, | |
| "learning_rate": 1.3506493506493508e-05, | |
| "loss": 0.0424, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.8125, | |
| "eval_accuracy": 0.95703125, | |
| "eval_loss": 0.1914406716823578, | |
| "eval_runtime": 5.2231, | |
| "eval_samples_per_second": 196.054, | |
| "eval_steps_per_second": 24.507, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0078125, | |
| "grad_norm": 0.06997977197170258, | |
| "learning_rate": 1.3294372294372295e-05, | |
| "loss": 0.031, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 8.203125, | |
| "grad_norm": 0.6767904162406921, | |
| "learning_rate": 1.3077922077922078e-05, | |
| "loss": 0.016, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 8.3984375, | |
| "grad_norm": 0.055935341864824295, | |
| "learning_rate": 1.2861471861471862e-05, | |
| "loss": 0.0128, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 8.59375, | |
| "grad_norm": 0.05341877415776253, | |
| "learning_rate": 1.2645021645021646e-05, | |
| "loss": 0.0123, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 8.7890625, | |
| "grad_norm": 0.051278505474328995, | |
| "learning_rate": 1.242857142857143e-05, | |
| "loss": 0.0243, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 8.984375, | |
| "grad_norm": 0.04905688017606735, | |
| "learning_rate": 1.2212121212121213e-05, | |
| "loss": 0.0116, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 9.1796875, | |
| "grad_norm": 0.05154794827103615, | |
| "learning_rate": 1.1995670995670996e-05, | |
| "loss": 0.0104, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "grad_norm": 0.04436648637056351, | |
| "learning_rate": 1.1779220779220779e-05, | |
| "loss": 0.0096, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.5703125, | |
| "grad_norm": 0.04592837765812874, | |
| "learning_rate": 1.1562770562770564e-05, | |
| "loss": 0.0096, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 9.765625, | |
| "grad_norm": 0.5957643985748291, | |
| "learning_rate": 1.1346320346320347e-05, | |
| "loss": 0.0108, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.765625, | |
| "eval_accuracy": 0.89453125, | |
| "eval_loss": 0.4534318447113037, | |
| "eval_runtime": 5.8562, | |
| "eval_samples_per_second": 174.857, | |
| "eval_steps_per_second": 21.857, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.9609375, | |
| "grad_norm": 0.3423336446285248, | |
| "learning_rate": 1.112987012987013e-05, | |
| "loss": 0.0183, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 10.15625, | |
| "grad_norm": 0.04016263037919998, | |
| "learning_rate": 1.0913419913419913e-05, | |
| "loss": 0.016, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 10.3515625, | |
| "grad_norm": 0.04166042432188988, | |
| "learning_rate": 1.0696969696969696e-05, | |
| "loss": 0.0077, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 10.546875, | |
| "grad_norm": 0.036698199808597565, | |
| "learning_rate": 1.0480519480519481e-05, | |
| "loss": 0.0077, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 10.7421875, | |
| "grad_norm": 0.03544982895255089, | |
| "learning_rate": 1.0264069264069264e-05, | |
| "loss": 0.0076, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 10.9375, | |
| "grad_norm": 0.033364105969667435, | |
| "learning_rate": 1.0047619047619048e-05, | |
| "loss": 0.0072, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 11.1328125, | |
| "grad_norm": 0.034085117280483246, | |
| "learning_rate": 9.831168831168832e-06, | |
| "loss": 0.007, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 11.328125, | |
| "grad_norm": 0.03140813857316971, | |
| "learning_rate": 9.614718614718616e-06, | |
| "loss": 0.0065, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 11.5234375, | |
| "grad_norm": 0.035150136798620224, | |
| "learning_rate": 9.398268398268399e-06, | |
| "loss": 0.014, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 11.71875, | |
| "grad_norm": 0.029343638569116592, | |
| "learning_rate": 9.181818181818184e-06, | |
| "loss": 0.0088, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.71875, | |
| "eval_accuracy": 0.9580078125, | |
| "eval_loss": 0.1553695797920227, | |
| "eval_runtime": 5.2716, | |
| "eval_samples_per_second": 194.25, | |
| "eval_steps_per_second": 24.281, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.9140625, | |
| "grad_norm": 0.02912583015859127, | |
| "learning_rate": 8.965367965367967e-06, | |
| "loss": 0.0063, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 12.109375, | |
| "grad_norm": 0.02827904000878334, | |
| "learning_rate": 8.74891774891775e-06, | |
| "loss": 0.0067, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 12.3046875, | |
| "grad_norm": 0.02774711512029171, | |
| "learning_rate": 8.536796536796537e-06, | |
| "loss": 0.0088, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 0.027258064597845078, | |
| "learning_rate": 8.32034632034632e-06, | |
| "loss": 0.0055, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 12.6953125, | |
| "grad_norm": 0.02587069943547249, | |
| "learning_rate": 8.103896103896105e-06, | |
| "loss": 0.0053, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 12.890625, | |
| "grad_norm": 0.02641950733959675, | |
| "learning_rate": 7.887445887445888e-06, | |
| "loss": 0.0054, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 13.0859375, | |
| "grad_norm": 0.028370605781674385, | |
| "learning_rate": 7.670995670995672e-06, | |
| "loss": 0.0052, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 13.28125, | |
| "grad_norm": 0.02395276166498661, | |
| "learning_rate": 7.454545454545456e-06, | |
| "loss": 0.0049, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 13.4765625, | |
| "grad_norm": 0.022906098514795303, | |
| "learning_rate": 7.238095238095239e-06, | |
| "loss": 0.0049, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 13.671875, | |
| "grad_norm": 0.023122085258364677, | |
| "learning_rate": 7.021645021645023e-06, | |
| "loss": 0.0051, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 13.671875, | |
| "eval_accuracy": 0.958984375, | |
| "eval_loss": 0.16660287976264954, | |
| "eval_runtime": 5.1728, | |
| "eval_samples_per_second": 197.959, | |
| "eval_steps_per_second": 24.745, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 13.8671875, | |
| "grad_norm": 0.022511709481477737, | |
| "learning_rate": 6.805194805194806e-06, | |
| "loss": 0.0047, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 14.0625, | |
| "grad_norm": 0.06705322116613388, | |
| "learning_rate": 6.588744588744589e-06, | |
| "loss": 0.0061, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 14.2578125, | |
| "grad_norm": 0.032343845814466476, | |
| "learning_rate": 6.372294372294373e-06, | |
| "loss": 0.0167, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 14.453125, | |
| "grad_norm": 0.040268197655677795, | |
| "learning_rate": 6.155844155844156e-06, | |
| "loss": 0.0059, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 14.6484375, | |
| "grad_norm": 0.020453229546546936, | |
| "learning_rate": 5.93939393939394e-06, | |
| "loss": 0.0044, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 14.84375, | |
| "grad_norm": 0.019762134179472923, | |
| "learning_rate": 5.722943722943723e-06, | |
| "loss": 0.0044, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 15.0390625, | |
| "grad_norm": 0.020851831883192062, | |
| "learning_rate": 5.5064935064935065e-06, | |
| "loss": 0.0041, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 15.234375, | |
| "grad_norm": 0.01954171247780323, | |
| "learning_rate": 5.2900432900432905e-06, | |
| "loss": 0.0041, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 15.4296875, | |
| "grad_norm": 0.020014379173517227, | |
| "learning_rate": 5.073593073593074e-06, | |
| "loss": 0.0042, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 15.625, | |
| "grad_norm": 0.01890181563794613, | |
| "learning_rate": 4.857142857142858e-06, | |
| "loss": 0.0039, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 15.625, | |
| "eval_accuracy": 0.96484375, | |
| "eval_loss": 0.15439653396606445, | |
| "eval_runtime": 5.142, | |
| "eval_samples_per_second": 199.146, | |
| "eval_steps_per_second": 24.893, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 15.8203125, | |
| "grad_norm": 0.017986547201871872, | |
| "learning_rate": 4.640692640692641e-06, | |
| "loss": 0.0038, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 16.015625, | |
| "grad_norm": 0.02049080841243267, | |
| "learning_rate": 4.424242424242425e-06, | |
| "loss": 0.0037, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 16.2109375, | |
| "grad_norm": 0.018345637246966362, | |
| "learning_rate": 4.207792207792208e-06, | |
| "loss": 0.0038, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 16.40625, | |
| "grad_norm": 0.018377136439085007, | |
| "learning_rate": 3.991341991341992e-06, | |
| "loss": 0.0037, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 16.6015625, | |
| "grad_norm": 0.01728520356118679, | |
| "learning_rate": 3.774891774891775e-06, | |
| "loss": 0.0037, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 16.796875, | |
| "grad_norm": 0.017723916098475456, | |
| "learning_rate": 3.558441558441559e-06, | |
| "loss": 0.0035, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 16.9921875, | |
| "grad_norm": 0.017535051330924034, | |
| "learning_rate": 3.3419913419913426e-06, | |
| "loss": 0.0036, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 17.1875, | |
| "grad_norm": 0.01705988124012947, | |
| "learning_rate": 3.1255411255411257e-06, | |
| "loss": 0.0036, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 17.3828125, | |
| "grad_norm": 0.01828162744641304, | |
| "learning_rate": 2.9090909090909093e-06, | |
| "loss": 0.0035, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 17.578125, | |
| "grad_norm": 0.026976697146892548, | |
| "learning_rate": 2.692640692640693e-06, | |
| "loss": 0.0034, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 17.578125, | |
| "eval_accuracy": 0.96484375, | |
| "eval_loss": 0.15748074650764465, | |
| "eval_runtime": 5.486, | |
| "eval_samples_per_second": 186.655, | |
| "eval_steps_per_second": 23.332, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 17.7734375, | |
| "grad_norm": 0.01693248189985752, | |
| "learning_rate": 2.4761904761904764e-06, | |
| "loss": 0.0033, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 17.96875, | |
| "grad_norm": 0.016402654349803925, | |
| "learning_rate": 2.25974025974026e-06, | |
| "loss": 0.0034, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 18.1640625, | |
| "grad_norm": 0.01663408987224102, | |
| "learning_rate": 2.0432900432900435e-06, | |
| "loss": 0.0033, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 18.359375, | |
| "grad_norm": 0.02396741323173046, | |
| "learning_rate": 1.8268398268398269e-06, | |
| "loss": 0.0034, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 18.5546875, | |
| "grad_norm": 0.016884848475456238, | |
| "learning_rate": 1.6103896103896105e-06, | |
| "loss": 0.0032, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "grad_norm": 0.016911398619413376, | |
| "learning_rate": 1.3939393939393942e-06, | |
| "loss": 0.0033, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 18.9453125, | |
| "grad_norm": 0.01589149609208107, | |
| "learning_rate": 1.1774891774891776e-06, | |
| "loss": 0.0033, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 19.140625, | |
| "grad_norm": 0.017035914584994316, | |
| "learning_rate": 9.610389610389612e-07, | |
| "loss": 0.0032, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 19.3359375, | |
| "grad_norm": 0.01621602661907673, | |
| "learning_rate": 7.445887445887446e-07, | |
| "loss": 0.0033, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 19.53125, | |
| "grad_norm": 0.015999143943190575, | |
| "learning_rate": 5.281385281385282e-07, | |
| "loss": 0.003, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 19.53125, | |
| "eval_accuracy": 0.9658203125, | |
| "eval_loss": 0.15924891829490662, | |
| "eval_runtime": 5.5508, | |
| "eval_samples_per_second": 184.477, | |
| "eval_steps_per_second": 23.06, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 19.7265625, | |
| "grad_norm": 0.014971093274652958, | |
| "learning_rate": 3.116883116883117e-07, | |
| "loss": 0.0033, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 19.921875, | |
| "grad_norm": 0.015599260106682777, | |
| "learning_rate": 9.523809523809525e-08, | |
| "loss": 0.0031, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 5120, | |
| "total_flos": 6.348257582037074e+18, | |
| "train_loss": 0.15641047188546509, | |
| "train_runtime": 1165.0601, | |
| "train_samples_per_second": 70.314, | |
| "train_steps_per_second": 4.395 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 5120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.348257582037074e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |