| { | |
| "best_metric": 0.8585118376550169, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-galaxy10-decals/checkpoint-2480", | |
| "epoch": 19.879759519038075, | |
| "eval_steps": 500, | |
| "global_step": 2480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08016032064128256, | |
| "grad_norm": 3.107008457183838, | |
| "learning_rate": 2.0161290322580646e-06, | |
| "loss": 2.3373, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16032064128256512, | |
| "grad_norm": 3.129793167114258, | |
| "learning_rate": 4.032258064516129e-06, | |
| "loss": 2.2991, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24048096192384769, | |
| "grad_norm": 3.2072296142578125, | |
| "learning_rate": 6.048387096774194e-06, | |
| "loss": 2.2401, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32064128256513025, | |
| "grad_norm": 3.172947406768799, | |
| "learning_rate": 8.064516129032258e-06, | |
| "loss": 2.1683, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.40080160320641284, | |
| "grad_norm": 4.017691612243652, | |
| "learning_rate": 1.0080645161290323e-05, | |
| "loss": 2.0858, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48096192384769537, | |
| "grad_norm": 4.591518402099609, | |
| "learning_rate": 1.2096774193548388e-05, | |
| "loss": 1.9855, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.561122244488978, | |
| "grad_norm": 6.409041404724121, | |
| "learning_rate": 1.4112903225806454e-05, | |
| "loss": 1.8467, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6412825651302605, | |
| "grad_norm": 7.812229633331299, | |
| "learning_rate": 1.6129032258064517e-05, | |
| "loss": 1.7127, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7214428857715431, | |
| "grad_norm": 9.216848373413086, | |
| "learning_rate": 1.8145161290322583e-05, | |
| "loss": 1.551, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8016032064128257, | |
| "grad_norm": 7.334836483001709, | |
| "learning_rate": 2.0161290322580645e-05, | |
| "loss": 1.4516, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8817635270541082, | |
| "grad_norm": 7.710146427154541, | |
| "learning_rate": 2.217741935483871e-05, | |
| "loss": 1.3589, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9619238476953907, | |
| "grad_norm": 11.563941955566406, | |
| "learning_rate": 2.4193548387096777e-05, | |
| "loss": 1.318, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9939879759519038, | |
| "eval_accuracy": 0.6358511837655016, | |
| "eval_loss": 1.0409355163574219, | |
| "eval_runtime": 18.6446, | |
| "eval_samples_per_second": 95.148, | |
| "eval_steps_per_second": 3.004, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 1.0420841683366733, | |
| "grad_norm": 9.838645935058594, | |
| "learning_rate": 2.620967741935484e-05, | |
| "loss": 1.2543, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.122244488977956, | |
| "grad_norm": 9.036652565002441, | |
| "learning_rate": 2.822580645161291e-05, | |
| "loss": 1.2318, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.2024048096192386, | |
| "grad_norm": 7.428995132446289, | |
| "learning_rate": 3.024193548387097e-05, | |
| "loss": 1.2297, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.282565130260521, | |
| "grad_norm": 7.845966815948486, | |
| "learning_rate": 3.2258064516129034e-05, | |
| "loss": 1.1232, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3627254509018036, | |
| "grad_norm": 7.774607181549072, | |
| "learning_rate": 3.427419354838709e-05, | |
| "loss": 1.0894, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.4428857715430863, | |
| "grad_norm": 12.192733764648438, | |
| "learning_rate": 3.6290322580645165e-05, | |
| "loss": 1.0552, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.5230460921843687, | |
| "grad_norm": 7.907654762268066, | |
| "learning_rate": 3.8306451612903224e-05, | |
| "loss": 1.0559, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.6032064128256514, | |
| "grad_norm": 8.524667739868164, | |
| "learning_rate": 4.032258064516129e-05, | |
| "loss": 1.0315, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6833667334669338, | |
| "grad_norm": 9.733327865600586, | |
| "learning_rate": 4.2338709677419356e-05, | |
| "loss": 1.0234, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.7635270541082164, | |
| "grad_norm": 8.974815368652344, | |
| "learning_rate": 4.435483870967742e-05, | |
| "loss": 0.9642, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.843687374749499, | |
| "grad_norm": 10.060734748840332, | |
| "learning_rate": 4.637096774193548e-05, | |
| "loss": 0.9358, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.9238476953907817, | |
| "grad_norm": 10.902128219604492, | |
| "learning_rate": 4.8387096774193554e-05, | |
| "loss": 0.9268, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9959919839679359, | |
| "eval_accuracy": 0.7497181510710259, | |
| "eval_loss": 0.7163556814193726, | |
| "eval_runtime": 15.1868, | |
| "eval_samples_per_second": 116.812, | |
| "eval_steps_per_second": 3.687, | |
| "step": 249 | |
| }, | |
| { | |
| "epoch": 2.004008016032064, | |
| "grad_norm": 8.283767700195312, | |
| "learning_rate": 4.995519713261649e-05, | |
| "loss": 0.915, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0841683366733466, | |
| "grad_norm": 9.61968994140625, | |
| "learning_rate": 4.973118279569893e-05, | |
| "loss": 0.8681, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.164328657314629, | |
| "grad_norm": 8.20541763305664, | |
| "learning_rate": 4.950716845878137e-05, | |
| "loss": 0.9177, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.244488977955912, | |
| "grad_norm": 6.9433369636535645, | |
| "learning_rate": 4.92831541218638e-05, | |
| "loss": 0.8946, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.3246492985971945, | |
| "grad_norm": 10.144634246826172, | |
| "learning_rate": 4.905913978494624e-05, | |
| "loss": 0.8933, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.404809619238477, | |
| "grad_norm": 12.519510269165039, | |
| "learning_rate": 4.8835125448028677e-05, | |
| "loss": 0.8515, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.4849699398797593, | |
| "grad_norm": 7.839648246765137, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.9103, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.565130260521042, | |
| "grad_norm": 9.056415557861328, | |
| "learning_rate": 4.8387096774193554e-05, | |
| "loss": 0.8601, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.6452905811623246, | |
| "grad_norm": 10.264932632446289, | |
| "learning_rate": 4.8163082437275986e-05, | |
| "loss": 0.8363, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.7254509018036073, | |
| "grad_norm": 8.888627052307129, | |
| "learning_rate": 4.7939068100358424e-05, | |
| "loss": 0.8075, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.80561122244489, | |
| "grad_norm": 10.737099647521973, | |
| "learning_rate": 4.771505376344086e-05, | |
| "loss": 0.9107, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.8857715430861726, | |
| "grad_norm": 7.076110363006592, | |
| "learning_rate": 4.74910394265233e-05, | |
| "loss": 0.7776, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.9659318637274548, | |
| "grad_norm": 7.52149772644043, | |
| "learning_rate": 4.726702508960574e-05, | |
| "loss": 0.8221, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.997995991983968, | |
| "eval_accuracy": 0.7874859075535513, | |
| "eval_loss": 0.6210038065910339, | |
| "eval_runtime": 16.893, | |
| "eval_samples_per_second": 105.014, | |
| "eval_steps_per_second": 3.315, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 3.0460921843687374, | |
| "grad_norm": 10.58385181427002, | |
| "learning_rate": 4.704301075268818e-05, | |
| "loss": 0.8115, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.12625250501002, | |
| "grad_norm": 11.055846214294434, | |
| "learning_rate": 4.681899641577061e-05, | |
| "loss": 0.7765, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.2064128256513027, | |
| "grad_norm": 8.763452529907227, | |
| "learning_rate": 4.659498207885305e-05, | |
| "loss": 0.7307, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.2865731462925853, | |
| "grad_norm": 6.804797172546387, | |
| "learning_rate": 4.637096774193548e-05, | |
| "loss": 0.7876, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.3667334669338675, | |
| "grad_norm": 6.112203121185303, | |
| "learning_rate": 4.614695340501792e-05, | |
| "loss": 0.7391, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.44689378757515, | |
| "grad_norm": 8.772920608520508, | |
| "learning_rate": 4.5922939068100365e-05, | |
| "loss": 0.7453, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.527054108216433, | |
| "grad_norm": 5.974344253540039, | |
| "learning_rate": 4.56989247311828e-05, | |
| "loss": 0.7345, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.6072144288577155, | |
| "grad_norm": 8.748202323913574, | |
| "learning_rate": 4.5474910394265236e-05, | |
| "loss": 0.8431, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.687374749498998, | |
| "grad_norm": 8.03186321258545, | |
| "learning_rate": 4.5250896057347674e-05, | |
| "loss": 0.7337, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.7675350701402808, | |
| "grad_norm": 7.393523216247559, | |
| "learning_rate": 4.5026881720430106e-05, | |
| "loss": 0.7451, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.847695390781563, | |
| "grad_norm": 6.694340705871582, | |
| "learning_rate": 4.4802867383512545e-05, | |
| "loss": 0.8017, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.9278557114228456, | |
| "grad_norm": 6.493546962738037, | |
| "learning_rate": 4.4578853046594983e-05, | |
| "loss": 0.7276, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8162344983089064, | |
| "eval_loss": 0.5563604235649109, | |
| "eval_runtime": 20.2707, | |
| "eval_samples_per_second": 87.515, | |
| "eval_steps_per_second": 2.763, | |
| "step": 499 | |
| }, | |
| { | |
| "epoch": 4.008016032064128, | |
| "grad_norm": 6.828517913818359, | |
| "learning_rate": 4.435483870967742e-05, | |
| "loss": 0.7295, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0881763527054105, | |
| "grad_norm": 8.60299301147461, | |
| "learning_rate": 4.413082437275986e-05, | |
| "loss": 0.7452, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.168336673346693, | |
| "grad_norm": 7.014772415161133, | |
| "learning_rate": 4.390681003584229e-05, | |
| "loss": 0.7108, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.248496993987976, | |
| "grad_norm": 10.6813325881958, | |
| "learning_rate": 4.368279569892473e-05, | |
| "loss": 0.7479, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.328657314629258, | |
| "grad_norm": 6.949085712432861, | |
| "learning_rate": 4.345878136200717e-05, | |
| "loss": 0.6518, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.408817635270541, | |
| "grad_norm": 6.594024658203125, | |
| "learning_rate": 4.323476702508961e-05, | |
| "loss": 0.7169, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.488977955911824, | |
| "grad_norm": 8.18333625793457, | |
| "learning_rate": 4.301075268817205e-05, | |
| "loss": 0.7185, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.569138276553106, | |
| "grad_norm": 5.988694190979004, | |
| "learning_rate": 4.2786738351254486e-05, | |
| "loss": 0.6685, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.649298597194389, | |
| "grad_norm": 7.308699607849121, | |
| "learning_rate": 4.256272401433692e-05, | |
| "loss": 0.6542, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.729458917835672, | |
| "grad_norm": 7.458045959472656, | |
| "learning_rate": 4.2338709677419356e-05, | |
| "loss": 0.6993, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.809619238476954, | |
| "grad_norm": 8.139283180236816, | |
| "learning_rate": 4.2114695340501795e-05, | |
| "loss": 0.7078, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.889779559118237, | |
| "grad_norm": 6.669909954071045, | |
| "learning_rate": 4.1890681003584233e-05, | |
| "loss": 0.6627, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.969939879759519, | |
| "grad_norm": 6.659294605255127, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.6425, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.993987975951904, | |
| "eval_accuracy": 0.8162344983089064, | |
| "eval_loss": 0.5226049423217773, | |
| "eval_runtime": 13.1794, | |
| "eval_samples_per_second": 134.604, | |
| "eval_steps_per_second": 4.249, | |
| "step": 623 | |
| }, | |
| { | |
| "epoch": 5.050100200400801, | |
| "grad_norm": 6.521961212158203, | |
| "learning_rate": 4.1442652329749104e-05, | |
| "loss": 0.6918, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.130260521042084, | |
| "grad_norm": 8.646223068237305, | |
| "learning_rate": 4.121863799283154e-05, | |
| "loss": 0.6712, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.210420841683367, | |
| "grad_norm": 5.398332118988037, | |
| "learning_rate": 4.099462365591398e-05, | |
| "loss": 0.6729, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.290581162324649, | |
| "grad_norm": 8.448481559753418, | |
| "learning_rate": 4.077060931899642e-05, | |
| "loss": 0.653, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.370741482965932, | |
| "grad_norm": 6.562283992767334, | |
| "learning_rate": 4.054659498207886e-05, | |
| "loss": 0.6835, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 5.4509018036072145, | |
| "grad_norm": 8.798486709594727, | |
| "learning_rate": 4.032258064516129e-05, | |
| "loss": 0.6849, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.531062124248497, | |
| "grad_norm": 8.284408569335938, | |
| "learning_rate": 4.009856630824373e-05, | |
| "loss": 0.6649, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 5.61122244488978, | |
| "grad_norm": 6.30844259262085, | |
| "learning_rate": 3.987455197132617e-05, | |
| "loss": 0.6893, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.6913827655310625, | |
| "grad_norm": 7.219119548797607, | |
| "learning_rate": 3.96505376344086e-05, | |
| "loss": 0.7007, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 5.771543086172345, | |
| "grad_norm": 8.133257865905762, | |
| "learning_rate": 3.9426523297491045e-05, | |
| "loss": 0.6595, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.851703406813627, | |
| "grad_norm": 7.415875434875488, | |
| "learning_rate": 3.9202508960573483e-05, | |
| "loss": 0.628, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.9318637274549095, | |
| "grad_norm": 7.713524341583252, | |
| "learning_rate": 3.8978494623655915e-05, | |
| "loss": 0.6518, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.995991983967936, | |
| "eval_accuracy": 0.818489289740699, | |
| "eval_loss": 0.537726879119873, | |
| "eval_runtime": 20.02, | |
| "eval_samples_per_second": 88.611, | |
| "eval_steps_per_second": 2.797, | |
| "step": 748 | |
| }, | |
| { | |
| "epoch": 6.012024048096192, | |
| "grad_norm": 7.509452819824219, | |
| "learning_rate": 3.8754480286738354e-05, | |
| "loss": 0.6262, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.092184368737475, | |
| "grad_norm": 6.618509769439697, | |
| "learning_rate": 3.8530465949820786e-05, | |
| "loss": 0.6559, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 6.1723446893787575, | |
| "grad_norm": 5.969357490539551, | |
| "learning_rate": 3.8306451612903224e-05, | |
| "loss": 0.6324, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.25250501002004, | |
| "grad_norm": 7.1045918464660645, | |
| "learning_rate": 3.808243727598566e-05, | |
| "loss": 0.5872, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.332665330661323, | |
| "grad_norm": 6.669059753417969, | |
| "learning_rate": 3.78584229390681e-05, | |
| "loss": 0.581, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 6.412825651302605, | |
| "grad_norm": 7.305534839630127, | |
| "learning_rate": 3.763440860215054e-05, | |
| "loss": 0.6319, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.492985971943888, | |
| "grad_norm": 5.811188697814941, | |
| "learning_rate": 3.741039426523298e-05, | |
| "loss": 0.6461, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 6.573146292585171, | |
| "grad_norm": 5.710335731506348, | |
| "learning_rate": 3.718637992831541e-05, | |
| "loss": 0.6673, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 6.653306613226453, | |
| "grad_norm": 8.094255447387695, | |
| "learning_rate": 3.696236559139785e-05, | |
| "loss": 0.6506, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 6.733466933867735, | |
| "grad_norm": 6.600334644317627, | |
| "learning_rate": 3.673835125448029e-05, | |
| "loss": 0.6437, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 6.813627254509018, | |
| "grad_norm": 7.381925106048584, | |
| "learning_rate": 3.651433691756273e-05, | |
| "loss": 0.6173, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 6.8937875751503, | |
| "grad_norm": 7.205611705780029, | |
| "learning_rate": 3.6290322580645165e-05, | |
| "loss": 0.6427, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.973947895791583, | |
| "grad_norm": 6.106593608856201, | |
| "learning_rate": 3.60663082437276e-05, | |
| "loss": 0.6096, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.997995991983968, | |
| "eval_accuracy": 0.8218714768883878, | |
| "eval_loss": 0.5341029167175293, | |
| "eval_runtime": 13.1967, | |
| "eval_samples_per_second": 134.427, | |
| "eval_steps_per_second": 4.243, | |
| "step": 873 | |
| }, | |
| { | |
| "epoch": 7.054108216432866, | |
| "grad_norm": 7.432844161987305, | |
| "learning_rate": 3.5842293906810036e-05, | |
| "loss": 0.6373, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 7.134268537074148, | |
| "grad_norm": 7.420022010803223, | |
| "learning_rate": 3.5618279569892474e-05, | |
| "loss": 0.6481, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 7.214428857715431, | |
| "grad_norm": 7.222751617431641, | |
| "learning_rate": 3.539426523297491e-05, | |
| "loss": 0.6267, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.294589178356714, | |
| "grad_norm": 7.050006866455078, | |
| "learning_rate": 3.517025089605735e-05, | |
| "loss": 0.6107, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.374749498997996, | |
| "grad_norm": 8.168829917907715, | |
| "learning_rate": 3.494623655913979e-05, | |
| "loss": 0.6008, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 7.454909819639279, | |
| "grad_norm": 5.229215145111084, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.6355, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 7.5350701402805615, | |
| "grad_norm": 6.677180290222168, | |
| "learning_rate": 3.449820788530466e-05, | |
| "loss": 0.6021, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 7.615230460921843, | |
| "grad_norm": 5.7284698486328125, | |
| "learning_rate": 3.427419354838709e-05, | |
| "loss": 0.602, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 7.695390781563126, | |
| "grad_norm": 7.613159656524658, | |
| "learning_rate": 3.405017921146954e-05, | |
| "loss": 0.578, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 7.775551102204409, | |
| "grad_norm": 7.990455150604248, | |
| "learning_rate": 3.382616487455198e-05, | |
| "loss": 0.6064, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 7.855711422845691, | |
| "grad_norm": 7.88253116607666, | |
| "learning_rate": 3.360215053763441e-05, | |
| "loss": 0.5796, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 7.935871743486974, | |
| "grad_norm": 6.4569091796875, | |
| "learning_rate": 3.337813620071685e-05, | |
| "loss": 0.6282, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8399098083427283, | |
| "eval_loss": 0.4718434512615204, | |
| "eval_runtime": 18.7441, | |
| "eval_samples_per_second": 94.643, | |
| "eval_steps_per_second": 2.988, | |
| "step": 998 | |
| }, | |
| { | |
| "epoch": 8.016032064128256, | |
| "grad_norm": 6.175160884857178, | |
| "learning_rate": 3.3154121863799286e-05, | |
| "loss": 0.5892, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.09619238476954, | |
| "grad_norm": 6.699339389801025, | |
| "learning_rate": 3.293010752688172e-05, | |
| "loss": 0.5914, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 8.176352705410821, | |
| "grad_norm": 7.189827919006348, | |
| "learning_rate": 3.270609318996416e-05, | |
| "loss": 0.5755, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 8.256513026052104, | |
| "grad_norm": 7.274308204650879, | |
| "learning_rate": 3.24820788530466e-05, | |
| "loss": 0.6348, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 8.336673346693386, | |
| "grad_norm": 5.570709228515625, | |
| "learning_rate": 3.2258064516129034e-05, | |
| "loss": 0.589, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.41683366733467, | |
| "grad_norm": 6.513092994689941, | |
| "learning_rate": 3.203405017921147e-05, | |
| "loss": 0.6085, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 8.496993987975952, | |
| "grad_norm": 6.354593753814697, | |
| "learning_rate": 3.1810035842293904e-05, | |
| "loss": 0.5925, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 8.577154308617235, | |
| "grad_norm": 6.1644392013549805, | |
| "learning_rate": 3.158602150537634e-05, | |
| "loss": 0.5287, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 8.657314629258517, | |
| "grad_norm": 7.383876800537109, | |
| "learning_rate": 3.136200716845878e-05, | |
| "loss": 0.6475, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 8.7374749498998, | |
| "grad_norm": 9.010411262512207, | |
| "learning_rate": 3.113799283154122e-05, | |
| "loss": 0.5934, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 8.817635270541082, | |
| "grad_norm": 5.401876449584961, | |
| "learning_rate": 3.091397849462366e-05, | |
| "loss": 0.5658, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.897795591182366, | |
| "grad_norm": 5.095533847808838, | |
| "learning_rate": 3.06899641577061e-05, | |
| "loss": 0.5791, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 8.977955911823647, | |
| "grad_norm": 6.6335248947143555, | |
| "learning_rate": 3.046594982078853e-05, | |
| "loss": 0.5394, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 8.993987975951903, | |
| "eval_accuracy": 0.8280721533258174, | |
| "eval_loss": 0.5112709999084473, | |
| "eval_runtime": 25.2314, | |
| "eval_samples_per_second": 70.309, | |
| "eval_steps_per_second": 2.219, | |
| "step": 1122 | |
| }, | |
| { | |
| "epoch": 9.05811623246493, | |
| "grad_norm": 6.276222229003906, | |
| "learning_rate": 3.024193548387097e-05, | |
| "loss": 0.586, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 9.138276553106213, | |
| "grad_norm": 5.882013320922852, | |
| "learning_rate": 3.0017921146953403e-05, | |
| "loss": 0.6203, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 9.218436873747494, | |
| "grad_norm": 7.025397777557373, | |
| "learning_rate": 2.979390681003584e-05, | |
| "loss": 0.5429, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 9.298597194388778, | |
| "grad_norm": 5.535187244415283, | |
| "learning_rate": 2.9569892473118284e-05, | |
| "loss": 0.5571, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 9.37875751503006, | |
| "grad_norm": 7.409646987915039, | |
| "learning_rate": 2.9345878136200715e-05, | |
| "loss": 0.5557, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.458917835671343, | |
| "grad_norm": 6.127359390258789, | |
| "learning_rate": 2.9121863799283154e-05, | |
| "loss": 0.5341, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 9.539078156312625, | |
| "grad_norm": 5.265384674072266, | |
| "learning_rate": 2.8897849462365596e-05, | |
| "loss": 0.5994, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 9.619238476953909, | |
| "grad_norm": 6.010611534118652, | |
| "learning_rate": 2.8673835125448028e-05, | |
| "loss": 0.5504, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.69939879759519, | |
| "grad_norm": 7.495913505554199, | |
| "learning_rate": 2.8449820788530467e-05, | |
| "loss": 0.5807, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 9.779559118236474, | |
| "grad_norm": 8.004411697387695, | |
| "learning_rate": 2.822580645161291e-05, | |
| "loss": 0.5911, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 9.859719438877756, | |
| "grad_norm": 7.179277420043945, | |
| "learning_rate": 2.800179211469534e-05, | |
| "loss": 0.5128, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 9.939879759519037, | |
| "grad_norm": 6.937490940093994, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.5718, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 9.995991983967937, | |
| "eval_accuracy": 0.8291995490417137, | |
| "eval_loss": 0.5018876791000366, | |
| "eval_runtime": 16.3385, | |
| "eval_samples_per_second": 108.578, | |
| "eval_steps_per_second": 3.427, | |
| "step": 1247 | |
| }, | |
| { | |
| "epoch": 10.02004008016032, | |
| "grad_norm": 7.811807632446289, | |
| "learning_rate": 2.7553763440860214e-05, | |
| "loss": 0.5727, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 10.100200400801603, | |
| "grad_norm": 7.441296577453613, | |
| "learning_rate": 2.7329749103942653e-05, | |
| "loss": 0.5742, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 10.180360721442886, | |
| "grad_norm": 6.93259334564209, | |
| "learning_rate": 2.710573476702509e-05, | |
| "loss": 0.4814, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 10.260521042084168, | |
| "grad_norm": 7.736974716186523, | |
| "learning_rate": 2.6881720430107527e-05, | |
| "loss": 0.5448, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 10.340681362725451, | |
| "grad_norm": 7.408446788787842, | |
| "learning_rate": 2.6657706093189965e-05, | |
| "loss": 0.5892, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 10.420841683366733, | |
| "grad_norm": 6.906106472015381, | |
| "learning_rate": 2.6433691756272404e-05, | |
| "loss": 0.5175, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.501002004008017, | |
| "grad_norm": 5.426215648651123, | |
| "learning_rate": 2.620967741935484e-05, | |
| "loss": 0.5977, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 10.581162324649299, | |
| "grad_norm": 5.591187477111816, | |
| "learning_rate": 2.5985663082437278e-05, | |
| "loss": 0.5157, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 10.661322645290582, | |
| "grad_norm": 7.416080474853516, | |
| "learning_rate": 2.5761648745519713e-05, | |
| "loss": 0.5578, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 10.741482965931864, | |
| "grad_norm": 6.815114498138428, | |
| "learning_rate": 2.5537634408602152e-05, | |
| "loss": 0.5387, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 10.821643286573146, | |
| "grad_norm": 8.681703567504883, | |
| "learning_rate": 2.531362007168459e-05, | |
| "loss": 0.5299, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 10.901803607214429, | |
| "grad_norm": 5.358316421508789, | |
| "learning_rate": 2.5089605734767026e-05, | |
| "loss": 0.59, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 10.98196392785571, | |
| "grad_norm": 8.894550323486328, | |
| "learning_rate": 2.4865591397849464e-05, | |
| "loss": 0.5507, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 10.997995991983968, | |
| "eval_accuracy": 0.8461104847801578, | |
| "eval_loss": 0.4545128643512726, | |
| "eval_runtime": 17.189, | |
| "eval_samples_per_second": 103.206, | |
| "eval_steps_per_second": 3.258, | |
| "step": 1372 | |
| }, | |
| { | |
| "epoch": 11.062124248496994, | |
| "grad_norm": 6.441153526306152, | |
| "learning_rate": 2.46415770609319e-05, | |
| "loss": 0.5485, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 11.142284569138276, | |
| "grad_norm": 5.337042808532715, | |
| "learning_rate": 2.4417562724014338e-05, | |
| "loss": 0.4986, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 11.22244488977956, | |
| "grad_norm": 5.681359767913818, | |
| "learning_rate": 2.4193548387096777e-05, | |
| "loss": 0.5213, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.302605210420841, | |
| "grad_norm": 6.7940778732299805, | |
| "learning_rate": 2.3969534050179212e-05, | |
| "loss": 0.5415, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 11.382765531062125, | |
| "grad_norm": 5.4109930992126465, | |
| "learning_rate": 2.374551971326165e-05, | |
| "loss": 0.5748, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 11.462925851703407, | |
| "grad_norm": 7.78901481628418, | |
| "learning_rate": 2.352150537634409e-05, | |
| "loss": 0.528, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 11.54308617234469, | |
| "grad_norm": 5.304915904998779, | |
| "learning_rate": 2.3297491039426525e-05, | |
| "loss": 0.5709, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 11.623246492985972, | |
| "grad_norm": 6.8759846687316895, | |
| "learning_rate": 2.307347670250896e-05, | |
| "loss": 0.5708, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 11.703406813627254, | |
| "grad_norm": 5.734496593475342, | |
| "learning_rate": 2.28494623655914e-05, | |
| "loss": 0.5178, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 11.783567134268537, | |
| "grad_norm": 7.169252395629883, | |
| "learning_rate": 2.2625448028673837e-05, | |
| "loss": 0.5595, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 11.863727454909819, | |
| "grad_norm": 6.391491413116455, | |
| "learning_rate": 2.2401433691756272e-05, | |
| "loss": 0.5883, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 11.943887775551103, | |
| "grad_norm": 5.931715965270996, | |
| "learning_rate": 2.217741935483871e-05, | |
| "loss": 0.4921, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8416009019165727, | |
| "eval_loss": 0.46128037571907043, | |
| "eval_runtime": 27.7579, | |
| "eval_samples_per_second": 63.91, | |
| "eval_steps_per_second": 2.017, | |
| "step": 1497 | |
| }, | |
| { | |
| "epoch": 12.024048096192384, | |
| "grad_norm": 5.848583221435547, | |
| "learning_rate": 2.1953405017921146e-05, | |
| "loss": 0.5388, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.104208416833668, | |
| "grad_norm": 5.273708343505859, | |
| "learning_rate": 2.1729390681003585e-05, | |
| "loss": 0.5259, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 12.18436873747495, | |
| "grad_norm": 6.022935390472412, | |
| "learning_rate": 2.1505376344086024e-05, | |
| "loss": 0.4928, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 12.264529058116233, | |
| "grad_norm": 4.965794563293457, | |
| "learning_rate": 2.128136200716846e-05, | |
| "loss": 0.537, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 12.344689378757515, | |
| "grad_norm": 6.983731746673584, | |
| "learning_rate": 2.1057347670250897e-05, | |
| "loss": 0.5258, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 12.424849699398798, | |
| "grad_norm": 6.290835380554199, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.5411, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 12.50501002004008, | |
| "grad_norm": 6.071152210235596, | |
| "learning_rate": 2.060931899641577e-05, | |
| "loss": 0.5421, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 12.585170340681362, | |
| "grad_norm": 7.99808931350708, | |
| "learning_rate": 2.038530465949821e-05, | |
| "loss": 0.5741, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 12.665330661322646, | |
| "grad_norm": 7.839056015014648, | |
| "learning_rate": 2.0161290322580645e-05, | |
| "loss": 0.5069, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 12.745490981963927, | |
| "grad_norm": 6.645950794219971, | |
| "learning_rate": 1.9937275985663084e-05, | |
| "loss": 0.5598, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 12.82565130260521, | |
| "grad_norm": 6.195275783538818, | |
| "learning_rate": 1.9713261648745522e-05, | |
| "loss": 0.5128, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.905811623246493, | |
| "grad_norm": 6.307319164276123, | |
| "learning_rate": 1.9489247311827958e-05, | |
| "loss": 0.5218, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 12.985971943887776, | |
| "grad_norm": 5.337151527404785, | |
| "learning_rate": 1.9265232974910393e-05, | |
| "loss": 0.5571, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 12.993987975951903, | |
| "eval_accuracy": 0.8416009019165727, | |
| "eval_loss": 0.45865094661712646, | |
| "eval_runtime": 13.0045, | |
| "eval_samples_per_second": 136.414, | |
| "eval_steps_per_second": 4.306, | |
| "step": 1621 | |
| }, | |
| { | |
| "epoch": 13.066132264529058, | |
| "grad_norm": 5.886476516723633, | |
| "learning_rate": 1.904121863799283e-05, | |
| "loss": 0.5145, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 13.146292585170341, | |
| "grad_norm": 7.12263298034668, | |
| "learning_rate": 1.881720430107527e-05, | |
| "loss": 0.4775, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 13.226452905811623, | |
| "grad_norm": 6.896437168121338, | |
| "learning_rate": 1.8593189964157705e-05, | |
| "loss": 0.4922, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 13.306613226452907, | |
| "grad_norm": 7.87682580947876, | |
| "learning_rate": 1.8369175627240144e-05, | |
| "loss": 0.5151, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 13.386773547094188, | |
| "grad_norm": 6.32350492477417, | |
| "learning_rate": 1.8145161290322583e-05, | |
| "loss": 0.5308, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 13.46693386773547, | |
| "grad_norm": 6.5004353523254395, | |
| "learning_rate": 1.7921146953405018e-05, | |
| "loss": 0.4914, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 13.547094188376754, | |
| "grad_norm": 6.300237655639648, | |
| "learning_rate": 1.7697132616487457e-05, | |
| "loss": 0.5254, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 13.627254509018035, | |
| "grad_norm": 6.251715660095215, | |
| "learning_rate": 1.7473118279569895e-05, | |
| "loss": 0.5197, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.707414829659319, | |
| "grad_norm": 8.49095630645752, | |
| "learning_rate": 1.724910394265233e-05, | |
| "loss": 0.5269, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 13.7875751503006, | |
| "grad_norm": 6.206210613250732, | |
| "learning_rate": 1.702508960573477e-05, | |
| "loss": 0.5317, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 13.867735470941884, | |
| "grad_norm": 6.281041145324707, | |
| "learning_rate": 1.6801075268817204e-05, | |
| "loss": 0.5212, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 13.947895791583166, | |
| "grad_norm": 5.863707542419434, | |
| "learning_rate": 1.6577060931899643e-05, | |
| "loss": 0.512, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 13.995991983967937, | |
| "eval_accuracy": 0.8511837655016911, | |
| "eval_loss": 0.46732643246650696, | |
| "eval_runtime": 20.8784, | |
| "eval_samples_per_second": 84.968, | |
| "eval_steps_per_second": 2.682, | |
| "step": 1746 | |
| }, | |
| { | |
| "epoch": 14.02805611222445, | |
| "grad_norm": 7.321824550628662, | |
| "learning_rate": 1.635304659498208e-05, | |
| "loss": 0.5023, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 14.108216432865731, | |
| "grad_norm": 7.074238300323486, | |
| "learning_rate": 1.6129032258064517e-05, | |
| "loss": 0.5645, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 14.188376753507015, | |
| "grad_norm": 4.364939212799072, | |
| "learning_rate": 1.5905017921146952e-05, | |
| "loss": 0.4858, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 14.268537074148297, | |
| "grad_norm": 6.330202102661133, | |
| "learning_rate": 1.568100358422939e-05, | |
| "loss": 0.5145, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 14.348697394789578, | |
| "grad_norm": 6.827199935913086, | |
| "learning_rate": 1.545698924731183e-05, | |
| "loss": 0.4995, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 14.428857715430862, | |
| "grad_norm": 5.992321014404297, | |
| "learning_rate": 1.5232974910394265e-05, | |
| "loss": 0.511, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 14.509018036072144, | |
| "grad_norm": 6.993434429168701, | |
| "learning_rate": 1.5008960573476701e-05, | |
| "loss": 0.535, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 14.589178356713427, | |
| "grad_norm": 6.39487886428833, | |
| "learning_rate": 1.4784946236559142e-05, | |
| "loss": 0.4754, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 14.669338677354709, | |
| "grad_norm": 7.741076946258545, | |
| "learning_rate": 1.4560931899641577e-05, | |
| "loss": 0.488, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 14.749498997995993, | |
| "grad_norm": 6.242033958435059, | |
| "learning_rate": 1.4336917562724014e-05, | |
| "loss": 0.4965, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 14.829659318637274, | |
| "grad_norm": 7.820639610290527, | |
| "learning_rate": 1.4112903225806454e-05, | |
| "loss": 0.5113, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 14.909819639278558, | |
| "grad_norm": 5.047755718231201, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.4945, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 14.98997995991984, | |
| "grad_norm": 6.295690059661865, | |
| "learning_rate": 1.3664874551971326e-05, | |
| "loss": 0.4855, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 14.997995991983968, | |
| "eval_accuracy": 0.8489289740698985, | |
| "eval_loss": 0.4640846252441406, | |
| "eval_runtime": 12.5307, | |
| "eval_samples_per_second": 141.573, | |
| "eval_steps_per_second": 4.469, | |
| "step": 1871 | |
| }, | |
| { | |
| "epoch": 15.070140280561123, | |
| "grad_norm": 5.256791114807129, | |
| "learning_rate": 1.3440860215053763e-05, | |
| "loss": 0.4476, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 15.150300601202405, | |
| "grad_norm": 6.767005920410156, | |
| "learning_rate": 1.3216845878136202e-05, | |
| "loss": 0.5136, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 15.230460921843687, | |
| "grad_norm": 6.730881690979004, | |
| "learning_rate": 1.2992831541218639e-05, | |
| "loss": 0.4952, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 15.31062124248497, | |
| "grad_norm": 5.721596717834473, | |
| "learning_rate": 1.2768817204301076e-05, | |
| "loss": 0.4984, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 15.390781563126252, | |
| "grad_norm": 5.367898941040039, | |
| "learning_rate": 1.2544802867383513e-05, | |
| "loss": 0.5366, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 15.470941883767535, | |
| "grad_norm": 7.774703502655029, | |
| "learning_rate": 1.232078853046595e-05, | |
| "loss": 0.4651, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 15.551102204408817, | |
| "grad_norm": 5.738451957702637, | |
| "learning_rate": 1.2096774193548388e-05, | |
| "loss": 0.4803, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 15.6312625250501, | |
| "grad_norm": 8.123086929321289, | |
| "learning_rate": 1.1872759856630825e-05, | |
| "loss": 0.5122, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 15.711422845691382, | |
| "grad_norm": 7.48280668258667, | |
| "learning_rate": 1.1648745519713262e-05, | |
| "loss": 0.4913, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 15.791583166332666, | |
| "grad_norm": 6.538034439086914, | |
| "learning_rate": 1.14247311827957e-05, | |
| "loss": 0.4653, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 15.871743486973948, | |
| "grad_norm": 6.203965663909912, | |
| "learning_rate": 1.1200716845878136e-05, | |
| "loss": 0.5026, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 15.951903807615231, | |
| "grad_norm": 7.2120490074157715, | |
| "learning_rate": 1.0976702508960573e-05, | |
| "loss": 0.4895, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8449830890642616, | |
| "eval_loss": 0.4555908143520355, | |
| "eval_runtime": 23.4028, | |
| "eval_samples_per_second": 75.803, | |
| "eval_steps_per_second": 2.393, | |
| "step": 1996 | |
| }, | |
| { | |
| "epoch": 16.03206412825651, | |
| "grad_norm": 5.893616199493408, | |
| "learning_rate": 1.0752688172043012e-05, | |
| "loss": 0.4819, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.112224448897795, | |
| "grad_norm": 7.0060133934021, | |
| "learning_rate": 1.0528673835125449e-05, | |
| "loss": 0.4928, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 16.19238476953908, | |
| "grad_norm": 5.684309005737305, | |
| "learning_rate": 1.0304659498207886e-05, | |
| "loss": 0.441, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 16.272545090180362, | |
| "grad_norm": 7.170827865600586, | |
| "learning_rate": 1.0080645161290323e-05, | |
| "loss": 0.4686, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 16.352705410821642, | |
| "grad_norm": 6.788947105407715, | |
| "learning_rate": 9.856630824372761e-06, | |
| "loss": 0.509, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 16.432865731462925, | |
| "grad_norm": 7.052069187164307, | |
| "learning_rate": 9.632616487455196e-06, | |
| "loss": 0.4727, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 16.51302605210421, | |
| "grad_norm": 6.445401668548584, | |
| "learning_rate": 9.408602150537635e-06, | |
| "loss": 0.5347, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 16.593186372745492, | |
| "grad_norm": 9.481761932373047, | |
| "learning_rate": 9.184587813620072e-06, | |
| "loss": 0.5089, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 16.673346693386772, | |
| "grad_norm": 5.852792739868164, | |
| "learning_rate": 8.960573476702509e-06, | |
| "loss": 0.4969, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 16.753507014028056, | |
| "grad_norm": 6.045396327972412, | |
| "learning_rate": 8.736559139784948e-06, | |
| "loss": 0.522, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 16.83366733466934, | |
| "grad_norm": 6.488787651062012, | |
| "learning_rate": 8.512544802867385e-06, | |
| "loss": 0.4813, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 16.91382765531062, | |
| "grad_norm": 7.611959934234619, | |
| "learning_rate": 8.288530465949821e-06, | |
| "loss": 0.4662, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 16.993987975951903, | |
| "grad_norm": 7.92677640914917, | |
| "learning_rate": 8.064516129032258e-06, | |
| "loss": 0.4809, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 16.993987975951903, | |
| "eval_accuracy": 0.8523111612175873, | |
| "eval_loss": 0.4317234456539154, | |
| "eval_runtime": 20.2324, | |
| "eval_samples_per_second": 87.681, | |
| "eval_steps_per_second": 2.768, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 17.074148296593187, | |
| "grad_norm": 6.190919876098633, | |
| "learning_rate": 7.840501792114695e-06, | |
| "loss": 0.4918, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 17.15430861723447, | |
| "grad_norm": 6.092956066131592, | |
| "learning_rate": 7.616487455197132e-06, | |
| "loss": 0.4602, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 17.23446893787575, | |
| "grad_norm": 7.813562870025635, | |
| "learning_rate": 7.392473118279571e-06, | |
| "loss": 0.5014, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 17.314629258517034, | |
| "grad_norm": 7.895810127258301, | |
| "learning_rate": 7.168458781362007e-06, | |
| "loss": 0.4704, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 17.394789579158317, | |
| "grad_norm": 7.144327640533447, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.5298, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 17.4749498997996, | |
| "grad_norm": 6.57069730758667, | |
| "learning_rate": 6.720430107526882e-06, | |
| "loss": 0.4797, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 17.55511022044088, | |
| "grad_norm": 5.174849510192871, | |
| "learning_rate": 6.4964157706093195e-06, | |
| "loss": 0.4697, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 17.635270541082164, | |
| "grad_norm": 4.489311218261719, | |
| "learning_rate": 6.2724014336917564e-06, | |
| "loss": 0.4764, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 17.715430861723448, | |
| "grad_norm": 8.91657829284668, | |
| "learning_rate": 6.048387096774194e-06, | |
| "loss": 0.4229, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 17.79559118236473, | |
| "grad_norm": 8.482898712158203, | |
| "learning_rate": 5.824372759856631e-06, | |
| "loss": 0.4835, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 17.87575150300601, | |
| "grad_norm": 7.158608436584473, | |
| "learning_rate": 5.600358422939068e-06, | |
| "loss": 0.4764, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 17.955911823647295, | |
| "grad_norm": 7.100325107574463, | |
| "learning_rate": 5.376344086021506e-06, | |
| "loss": 0.4785, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 17.995991983967937, | |
| "eval_accuracy": 0.8534385569334837, | |
| "eval_loss": 0.4337688088417053, | |
| "eval_runtime": 14.1747, | |
| "eval_samples_per_second": 125.152, | |
| "eval_steps_per_second": 3.951, | |
| "step": 2245 | |
| }, | |
| { | |
| "epoch": 18.03607214428858, | |
| "grad_norm": 6.301604747772217, | |
| "learning_rate": 5.152329749103943e-06, | |
| "loss": 0.4679, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 18.11623246492986, | |
| "grad_norm": 8.363497734069824, | |
| "learning_rate": 4.928315412186381e-06, | |
| "loss": 0.4164, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 18.196392785571142, | |
| "grad_norm": 6.71609354019165, | |
| "learning_rate": 4.7043010752688175e-06, | |
| "loss": 0.4596, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 18.276553106212425, | |
| "grad_norm": 5.690964221954346, | |
| "learning_rate": 4.4802867383512545e-06, | |
| "loss": 0.4774, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 18.35671342685371, | |
| "grad_norm": 7.212980270385742, | |
| "learning_rate": 4.256272401433692e-06, | |
| "loss": 0.5049, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 18.43687374749499, | |
| "grad_norm": 7.2161149978637695, | |
| "learning_rate": 4.032258064516129e-06, | |
| "loss": 0.4504, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 18.517034068136272, | |
| "grad_norm": 5.582963466644287, | |
| "learning_rate": 3.808243727598566e-06, | |
| "loss": 0.4624, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 18.597194388777556, | |
| "grad_norm": 6.577459335327148, | |
| "learning_rate": 3.5842293906810035e-06, | |
| "loss": 0.4636, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 18.677354709418836, | |
| "grad_norm": 6.8889594078063965, | |
| "learning_rate": 3.360215053763441e-06, | |
| "loss": 0.4418, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 18.75751503006012, | |
| "grad_norm": 7.460567951202393, | |
| "learning_rate": 3.1362007168458782e-06, | |
| "loss": 0.49, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 18.837675350701403, | |
| "grad_norm": 6.316689491271973, | |
| "learning_rate": 2.9121863799283156e-06, | |
| "loss": 0.4392, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 18.917835671342687, | |
| "grad_norm": 7.855792999267578, | |
| "learning_rate": 2.688172043010753e-06, | |
| "loss": 0.4779, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 18.997995991983966, | |
| "grad_norm": 5.641603946685791, | |
| "learning_rate": 2.4641577060931903e-06, | |
| "loss": 0.444, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 18.997995991983966, | |
| "eval_accuracy": 0.8579481397970687, | |
| "eval_loss": 0.4356663227081299, | |
| "eval_runtime": 12.8983, | |
| "eval_samples_per_second": 137.538, | |
| "eval_steps_per_second": 4.342, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 19.07815631262525, | |
| "grad_norm": 7.950003623962402, | |
| "learning_rate": 2.2401433691756272e-06, | |
| "loss": 0.4587, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 19.158316633266534, | |
| "grad_norm": 6.9599995613098145, | |
| "learning_rate": 2.0161290322580646e-06, | |
| "loss": 0.4754, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 19.238476953907817, | |
| "grad_norm": 6.827354907989502, | |
| "learning_rate": 1.7921146953405017e-06, | |
| "loss": 0.4576, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 19.318637274549097, | |
| "grad_norm": 5.975595951080322, | |
| "learning_rate": 1.5681003584229391e-06, | |
| "loss": 0.427, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 19.39879759519038, | |
| "grad_norm": 7.829305648803711, | |
| "learning_rate": 1.3440860215053765e-06, | |
| "loss": 0.5141, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 19.478957915831664, | |
| "grad_norm": 5.588257312774658, | |
| "learning_rate": 1.1200716845878136e-06, | |
| "loss": 0.4609, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 19.559118236472948, | |
| "grad_norm": 8.075860023498535, | |
| "learning_rate": 8.960573476702509e-07, | |
| "loss": 0.4977, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 19.639278557114228, | |
| "grad_norm": 7.977848052978516, | |
| "learning_rate": 6.720430107526882e-07, | |
| "loss": 0.4329, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 19.71943887775551, | |
| "grad_norm": 7.050076961517334, | |
| "learning_rate": 4.4802867383512544e-07, | |
| "loss": 0.4613, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 19.799599198396795, | |
| "grad_norm": 6.357409954071045, | |
| "learning_rate": 2.2401433691756272e-07, | |
| "loss": 0.4457, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 19.879759519038075, | |
| "grad_norm": 4.937966346740723, | |
| "learning_rate": 0.0, | |
| "loss": 0.4255, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 19.879759519038075, | |
| "eval_accuracy": 0.8585118376550169, | |
| "eval_loss": 0.4356611371040344, | |
| "eval_runtime": 26.523, | |
| "eval_samples_per_second": 66.885, | |
| "eval_steps_per_second": 2.111, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 19.879759519038075, | |
| "step": 2480, | |
| "total_flos": 1.0326291224762253e+19, | |
| "train_loss": 0.6660777115052746, | |
| "train_runtime": 7387.7979, | |
| "train_samples_per_second": 43.212, | |
| "train_steps_per_second": 0.336 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.0326291224762253e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |