| { | |
| "best_metric": 0.33602150537634407, | |
| "best_model_checkpoint": "vivit-b-16x2-kinetics400-finetuned-elder/checkpoint-145", | |
| "epoch": 3.2447916666666665, | |
| "eval_steps": 500, | |
| "global_step": 576, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 10.964410781860352, | |
| "learning_rate": 8.620689655172414e-06, | |
| "loss": 1.905, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 9.987532615661621, | |
| "learning_rate": 1.7241379310344828e-05, | |
| "loss": 1.8323, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 9.509833335876465, | |
| "learning_rate": 2.5862068965517244e-05, | |
| "loss": 1.749, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 10.654390335083008, | |
| "learning_rate": 3.4482758620689657e-05, | |
| "loss": 1.6316, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 10.157491683959961, | |
| "learning_rate": 4.3103448275862066e-05, | |
| "loss": 1.7427, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 9.177325248718262, | |
| "learning_rate": 4.980694980694981e-05, | |
| "loss": 1.6621, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 8.330065727233887, | |
| "learning_rate": 4.884169884169885e-05, | |
| "loss": 1.771, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 8.352508544921875, | |
| "learning_rate": 4.787644787644788e-05, | |
| "loss": 1.6952, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 7.394218921661377, | |
| "learning_rate": 4.6911196911196914e-05, | |
| "loss": 1.5736, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 6.307917594909668, | |
| "learning_rate": 4.594594594594595e-05, | |
| "loss": 1.6483, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 5.816230773925781, | |
| "learning_rate": 4.498069498069498e-05, | |
| "loss": 1.7185, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 8.665311813354492, | |
| "learning_rate": 4.401544401544402e-05, | |
| "loss": 1.6222, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 6.528253555297852, | |
| "learning_rate": 4.305019305019305e-05, | |
| "loss": 1.6889, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 5.285109996795654, | |
| "learning_rate": 4.2084942084942086e-05, | |
| "loss": 1.6183, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_accuracy": 0.33602150537634407, | |
| "eval_loss": 1.6138807535171509, | |
| "eval_runtime": 1859.4434, | |
| "eval_samples_per_second": 0.4, | |
| "eval_steps_per_second": 0.05, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 8.135934829711914, | |
| "learning_rate": 4.111969111969112e-05, | |
| "loss": 1.5727, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 6.669942378997803, | |
| "learning_rate": 4.015444015444015e-05, | |
| "loss": 1.6046, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 7.650122165679932, | |
| "learning_rate": 3.918918918918919e-05, | |
| "loss": 1.5688, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 6.249934196472168, | |
| "learning_rate": 3.822393822393823e-05, | |
| "loss": 1.7691, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 11.113762855529785, | |
| "learning_rate": 3.725868725868726e-05, | |
| "loss": 1.6297, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 8.34766960144043, | |
| "learning_rate": 3.6293436293436295e-05, | |
| "loss": 1.6644, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 7.016932964324951, | |
| "learning_rate": 3.532818532818533e-05, | |
| "loss": 1.6115, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 5.9432759284973145, | |
| "learning_rate": 3.436293436293436e-05, | |
| "loss": 1.5386, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 7.149511814117432, | |
| "learning_rate": 3.33976833976834e-05, | |
| "loss": 1.5963, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 6.586450576782227, | |
| "learning_rate": 3.2432432432432436e-05, | |
| "loss": 1.5814, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 6.453098297119141, | |
| "learning_rate": 3.1467181467181466e-05, | |
| "loss": 1.486, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 6.065974235534668, | |
| "learning_rate": 3.0501930501930504e-05, | |
| "loss": 1.5232, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 4.901158332824707, | |
| "learning_rate": 2.953667953667954e-05, | |
| "loss": 1.6137, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 8.322129249572754, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 1.5236, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 7.243752956390381, | |
| "learning_rate": 2.7606177606177608e-05, | |
| "loss": 1.5777, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 0.3024193548387097, | |
| "eval_loss": 1.6060892343521118, | |
| "eval_runtime": 1835.4134, | |
| "eval_samples_per_second": 0.405, | |
| "eval_steps_per_second": 0.051, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 7.097980499267578, | |
| "learning_rate": 2.6640926640926645e-05, | |
| "loss": 1.4382, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 5.670882225036621, | |
| "learning_rate": 2.5675675675675675e-05, | |
| "loss": 1.5916, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 6.845547199249268, | |
| "learning_rate": 2.4710424710424712e-05, | |
| "loss": 1.5396, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 8.554476737976074, | |
| "learning_rate": 2.3745173745173746e-05, | |
| "loss": 1.4087, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 10.264742851257324, | |
| "learning_rate": 2.277992277992278e-05, | |
| "loss": 1.524, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 9.64910888671875, | |
| "learning_rate": 2.1814671814671817e-05, | |
| "loss": 1.4529, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 4.2720723152160645, | |
| "learning_rate": 2.084942084942085e-05, | |
| "loss": 1.5956, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 7.2199811935424805, | |
| "learning_rate": 1.9884169884169884e-05, | |
| "loss": 1.5471, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 7.5855021476745605, | |
| "learning_rate": 1.891891891891892e-05, | |
| "loss": 1.3987, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 6.013571739196777, | |
| "learning_rate": 1.7953667953667955e-05, | |
| "loss": 1.4669, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 5.8329901695251465, | |
| "learning_rate": 1.698841698841699e-05, | |
| "loss": 1.5178, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 5.117478847503662, | |
| "learning_rate": 1.6023166023166026e-05, | |
| "loss": 1.4729, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 11.459242820739746, | |
| "learning_rate": 1.505791505791506e-05, | |
| "loss": 1.7009, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 6.949310302734375, | |
| "learning_rate": 1.4092664092664093e-05, | |
| "loss": 1.36, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_accuracy": 0.2862903225806452, | |
| "eval_loss": 1.644209623336792, | |
| "eval_runtime": 1825.2929, | |
| "eval_samples_per_second": 0.408, | |
| "eval_steps_per_second": 0.051, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 12.649101257324219, | |
| "learning_rate": 1.3127413127413127e-05, | |
| "loss": 1.4907, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 9.230242729187012, | |
| "learning_rate": 1.2162162162162164e-05, | |
| "loss": 1.3937, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 14.451587677001953, | |
| "learning_rate": 1.1196911196911197e-05, | |
| "loss": 1.4505, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 6.578619956970215, | |
| "learning_rate": 1.0231660231660233e-05, | |
| "loss": 1.3963, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 5.722152233123779, | |
| "learning_rate": 9.266409266409266e-06, | |
| "loss": 1.4609, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 6.25828742980957, | |
| "learning_rate": 8.301158301158302e-06, | |
| "loss": 1.3169, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "grad_norm": 10.180989265441895, | |
| "learning_rate": 7.335907335907337e-06, | |
| "loss": 1.3103, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 12.603837013244629, | |
| "learning_rate": 6.370656370656371e-06, | |
| "loss": 1.4784, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 9.059818267822266, | |
| "learning_rate": 5.405405405405406e-06, | |
| "loss": 1.3549, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 7.526132583618164, | |
| "learning_rate": 4.4401544401544405e-06, | |
| "loss": 1.5093, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "grad_norm": 6.272554397583008, | |
| "learning_rate": 3.474903474903475e-06, | |
| "loss": 1.3441, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 7.645261287689209, | |
| "learning_rate": 2.5096525096525096e-06, | |
| "loss": 1.5137, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "grad_norm": 6.750068187713623, | |
| "learning_rate": 1.5444015444015445e-06, | |
| "loss": 1.422, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 9.551359176635742, | |
| "learning_rate": 5.791505791505791e-07, | |
| "loss": 1.5395, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "eval_accuracy": 0.26881720430107525, | |
| "eval_loss": 1.6517891883850098, | |
| "eval_runtime": 1853.8749, | |
| "eval_samples_per_second": 0.401, | |
| "eval_steps_per_second": 0.05, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "step": 576, | |
| "total_flos": 1.178326479032982e+19, | |
| "train_loss": 1.5556091964244843, | |
| "train_runtime": 30077.0684, | |
| "train_samples_per_second": 0.153, | |
| "train_steps_per_second": 0.019 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "eval_accuracy": 0.32051282051282054, | |
| "eval_loss": 1.680675983428955, | |
| "eval_runtime": 1526.7516, | |
| "eval_samples_per_second": 0.409, | |
| "eval_steps_per_second": 0.051, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "eval_accuracy": 0.32051282051282054, | |
| "eval_loss": 1.680675983428955, | |
| "eval_runtime": 1508.1494, | |
| "eval_samples_per_second": 0.414, | |
| "eval_steps_per_second": 0.052, | |
| "step": 576 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 576, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 1.178326479032982e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |