| { | |
| "best_metric": 0.9992091735863978, | |
| "best_model_checkpoint": "vit-base-patch16-224-clothes-filter/checkpoint-1265", | |
| "epoch": 4.996047430830039, | |
| "eval_steps": 500, | |
| "global_step": 1580, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 4.780308246612549, | |
| "learning_rate": 3.1645569620253167e-06, | |
| "loss": 0.6282, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 4.528079986572266, | |
| "learning_rate": 6.329113924050633e-06, | |
| "loss": 0.5794, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 3.32427978515625, | |
| "learning_rate": 9.49367088607595e-06, | |
| "loss": 0.4682, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 3.10469388961792, | |
| "learning_rate": 1.2658227848101267e-05, | |
| "loss": 0.3436, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.8978112936019897, | |
| "learning_rate": 1.5822784810126583e-05, | |
| "loss": 0.2182, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 1.4250308275222778, | |
| "learning_rate": 1.89873417721519e-05, | |
| "loss": 0.1309, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 1.0968660116195679, | |
| "learning_rate": 2.2151898734177217e-05, | |
| "loss": 0.0826, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 2.002328872680664, | |
| "learning_rate": 2.5316455696202533e-05, | |
| "loss": 0.0477, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.20249253511428833, | |
| "learning_rate": 2.848101265822785e-05, | |
| "loss": 0.0207, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.15333689749240875, | |
| "learning_rate": 3.1645569620253167e-05, | |
| "loss": 0.0164, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.08237770944833755, | |
| "learning_rate": 3.4810126582278487e-05, | |
| "loss": 0.0419, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.6087208986282349, | |
| "learning_rate": 3.79746835443038e-05, | |
| "loss": 0.0081, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.059435680508613586, | |
| "learning_rate": 4.113924050632912e-05, | |
| "loss": 0.007, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.08203254640102386, | |
| "learning_rate": 4.430379746835443e-05, | |
| "loss": 0.022, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.059913892298936844, | |
| "learning_rate": 4.7468354430379746e-05, | |
| "loss": 0.02, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.04219400882720947, | |
| "learning_rate": 4.9929676511955e-05, | |
| "loss": 0.0241, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.4119158089160919, | |
| "learning_rate": 4.957805907172996e-05, | |
| "loss": 0.011, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.08845602720975876, | |
| "learning_rate": 4.9226441631504925e-05, | |
| "loss": 0.012, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.022793015465140343, | |
| "learning_rate": 4.887482419127989e-05, | |
| "loss": 0.0044, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.1266743242740631, | |
| "learning_rate": 4.852320675105486e-05, | |
| "loss": 0.0026, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.2308553010225296, | |
| "learning_rate": 4.817158931082982e-05, | |
| "loss": 0.0088, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.2685852348804474, | |
| "learning_rate": 4.7819971870604783e-05, | |
| "loss": 0.0191, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.012726900167763233, | |
| "learning_rate": 4.7468354430379746e-05, | |
| "loss": 0.0084, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.30946388840675354, | |
| "learning_rate": 4.7116736990154716e-05, | |
| "loss": 0.0242, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.03251117840409279, | |
| "learning_rate": 4.676511954992968e-05, | |
| "loss": 0.0028, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.622480034828186, | |
| "learning_rate": 4.641350210970464e-05, | |
| "loss": 0.0123, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.013131936080753803, | |
| "learning_rate": 4.606188466947961e-05, | |
| "loss": 0.0013, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.042751457542181015, | |
| "learning_rate": 4.5710267229254575e-05, | |
| "loss": 0.0403, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.023626720532774925, | |
| "learning_rate": 4.535864978902954e-05, | |
| "loss": 0.0376, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.03286024183034897, | |
| "learning_rate": 4.50070323488045e-05, | |
| "loss": 0.0095, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.19246149063110352, | |
| "learning_rate": 4.465541490857947e-05, | |
| "loss": 0.0462, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9936733886911823, | |
| "eval_loss": 0.016152312979102135, | |
| "eval_runtime": 16.8828, | |
| "eval_samples_per_second": 149.798, | |
| "eval_steps_per_second": 18.777, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 4.946556568145752, | |
| "learning_rate": 4.430379746835443e-05, | |
| "loss": 0.02, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 1.085129737854004, | |
| "learning_rate": 4.3952180028129396e-05, | |
| "loss": 0.0067, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.026508348062634468, | |
| "learning_rate": 4.3600562587904366e-05, | |
| "loss": 0.0151, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 0.13778889179229736, | |
| "learning_rate": 4.324894514767933e-05, | |
| "loss": 0.007, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 3.9214696884155273, | |
| "learning_rate": 4.289732770745429e-05, | |
| "loss": 0.0384, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 0.047678280621767044, | |
| "learning_rate": 4.2545710267229255e-05, | |
| "loss": 0.001, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.579052448272705, | |
| "learning_rate": 4.2194092827004224e-05, | |
| "loss": 0.0254, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 0.04058700054883957, | |
| "learning_rate": 4.184247538677919e-05, | |
| "loss": 0.0037, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 0.02530195750296116, | |
| "learning_rate": 4.149085794655415e-05, | |
| "loss": 0.0018, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 2.1537139415740967, | |
| "learning_rate": 4.113924050632912e-05, | |
| "loss": 0.0046, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 0.4535873234272003, | |
| "learning_rate": 4.078762306610408e-05, | |
| "loss": 0.006, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 0.050164151936769485, | |
| "learning_rate": 4.0436005625879046e-05, | |
| "loss": 0.0055, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 0.015650559216737747, | |
| "learning_rate": 4.008438818565401e-05, | |
| "loss": 0.0159, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 0.02410263940691948, | |
| "learning_rate": 3.973277074542898e-05, | |
| "loss": 0.0008, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.005263881757855415, | |
| "learning_rate": 3.938115330520394e-05, | |
| "loss": 0.004, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 0.005594250746071339, | |
| "learning_rate": 3.9029535864978904e-05, | |
| "loss": 0.0004, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.008686939254403114, | |
| "learning_rate": 3.867791842475387e-05, | |
| "loss": 0.0014, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.00374673493206501, | |
| "learning_rate": 3.832630098452884e-05, | |
| "loss": 0.0086, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 0.004833437968045473, | |
| "learning_rate": 3.79746835443038e-05, | |
| "loss": 0.0136, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 2.200552463531494, | |
| "learning_rate": 3.762306610407876e-05, | |
| "loss": 0.0156, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 0.1714821308851242, | |
| "learning_rate": 3.727144866385373e-05, | |
| "loss": 0.0006, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 0.009503379464149475, | |
| "learning_rate": 3.6919831223628695e-05, | |
| "loss": 0.0142, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 0.004495191853493452, | |
| "learning_rate": 3.656821378340366e-05, | |
| "loss": 0.0119, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 0.2889450192451477, | |
| "learning_rate": 3.621659634317862e-05, | |
| "loss": 0.0021, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 0.04732218757271767, | |
| "learning_rate": 3.586497890295359e-05, | |
| "loss": 0.005, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.710472047328949, | |
| "learning_rate": 3.551336146272855e-05, | |
| "loss": 0.0009, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 0.025769829750061035, | |
| "learning_rate": 3.516174402250352e-05, | |
| "loss": 0.0013, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 5.155148983001709, | |
| "learning_rate": 3.4810126582278487e-05, | |
| "loss": 0.016, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.009949293918907642, | |
| "learning_rate": 3.445850914205345e-05, | |
| "loss": 0.0005, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 0.004877807106822729, | |
| "learning_rate": 3.410689170182841e-05, | |
| "loss": 0.006, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 3.116375684738159, | |
| "learning_rate": 3.3755274261603375e-05, | |
| "loss": 0.0076, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 0.0024900627322494984, | |
| "learning_rate": 3.3403656821378345e-05, | |
| "loss": 0.0108, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9984183471727955, | |
| "eval_loss": 0.004202807787805796, | |
| "eval_runtime": 17.1708, | |
| "eval_samples_per_second": 147.285, | |
| "eval_steps_per_second": 18.462, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 0.027852624654769897, | |
| "learning_rate": 3.305203938115331e-05, | |
| "loss": 0.0126, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 0.00895751640200615, | |
| "learning_rate": 3.270042194092827e-05, | |
| "loss": 0.0247, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 0.005895175039768219, | |
| "learning_rate": 3.234880450070324e-05, | |
| "loss": 0.0031, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 0.0067118131555616856, | |
| "learning_rate": 3.1997187060478204e-05, | |
| "loss": 0.0123, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 3.036545991897583, | |
| "learning_rate": 3.1645569620253167e-05, | |
| "loss": 0.0139, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 0.013006511144340038, | |
| "learning_rate": 3.129395218002813e-05, | |
| "loss": 0.0007, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 0.011808563023805618, | |
| "learning_rate": 3.09423347398031e-05, | |
| "loss": 0.001, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.0029538481030613184, | |
| "learning_rate": 3.059071729957806e-05, | |
| "loss": 0.004, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 0.0031332324724644423, | |
| "learning_rate": 3.0239099859353025e-05, | |
| "loss": 0.0006, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 0.07193127274513245, | |
| "learning_rate": 2.9887482419127988e-05, | |
| "loss": 0.0134, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 0.009125332348048687, | |
| "learning_rate": 2.9535864978902954e-05, | |
| "loss": 0.0009, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 0.03563140332698822, | |
| "learning_rate": 2.9184247538677924e-05, | |
| "loss": 0.008, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.4033479690551758, | |
| "learning_rate": 2.8832630098452884e-05, | |
| "loss": 0.0006, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 0.005182890221476555, | |
| "learning_rate": 2.848101265822785e-05, | |
| "loss": 0.0003, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 0.06785815209150314, | |
| "learning_rate": 2.8129395218002813e-05, | |
| "loss": 0.0005, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.00515472237020731, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0116, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 0.06535543501377106, | |
| "learning_rate": 2.7426160337552742e-05, | |
| "loss": 0.0002, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.2995426654815674, | |
| "learning_rate": 2.707454289732771e-05, | |
| "loss": 0.014, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 0.011195150204002857, | |
| "learning_rate": 2.672292545710267e-05, | |
| "loss": 0.0257, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 0.00852285698056221, | |
| "learning_rate": 2.6371308016877638e-05, | |
| "loss": 0.0249, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 0.3155508041381836, | |
| "learning_rate": 2.6019690576652604e-05, | |
| "loss": 0.0008, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 0.00613983441144228, | |
| "learning_rate": 2.5668073136427567e-05, | |
| "loss": 0.0024, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 0.18053768575191498, | |
| "learning_rate": 2.5316455696202533e-05, | |
| "loss": 0.0021, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 0.3204176127910614, | |
| "learning_rate": 2.49648382559775e-05, | |
| "loss": 0.0087, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 0.019261712208390236, | |
| "learning_rate": 2.4613220815752462e-05, | |
| "loss": 0.0004, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 0.04258158802986145, | |
| "learning_rate": 2.426160337552743e-05, | |
| "loss": 0.0157, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.8219675421714783, | |
| "learning_rate": 2.3909985935302392e-05, | |
| "loss": 0.0018, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.007423636503517628, | |
| "learning_rate": 2.3558368495077358e-05, | |
| "loss": 0.0004, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 0.039796918630599976, | |
| "learning_rate": 2.320675105485232e-05, | |
| "loss": 0.0005, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 0.007585118990391493, | |
| "learning_rate": 2.2855133614627287e-05, | |
| "loss": 0.0408, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 0.004360842052847147, | |
| "learning_rate": 2.250351617440225e-05, | |
| "loss": 0.0051, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9972321075523922, | |
| "eval_loss": 0.0071137435734272, | |
| "eval_runtime": 16.9966, | |
| "eval_samples_per_second": 148.794, | |
| "eval_steps_per_second": 18.651, | |
| "step": 948 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.012101028114557266, | |
| "learning_rate": 2.2151898734177217e-05, | |
| "loss": 0.0003, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 0.021550431847572327, | |
| "learning_rate": 2.1800281293952183e-05, | |
| "loss": 0.0036, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 0.003491973737254739, | |
| "learning_rate": 2.1448663853727146e-05, | |
| "loss": 0.0002, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 0.0050508989952504635, | |
| "learning_rate": 2.1097046413502112e-05, | |
| "loss": 0.0007, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 0.004054068587720394, | |
| "learning_rate": 2.0745428973277075e-05, | |
| "loss": 0.0003, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 0.0022561594378203154, | |
| "learning_rate": 2.039381153305204e-05, | |
| "loss": 0.0007, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "grad_norm": 0.016699934378266335, | |
| "learning_rate": 2.0042194092827004e-05, | |
| "loss": 0.0052, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 0.016947178170084953, | |
| "learning_rate": 1.969057665260197e-05, | |
| "loss": 0.0027, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "grad_norm": 0.0021668283734470606, | |
| "learning_rate": 1.9338959212376934e-05, | |
| "loss": 0.0085, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "grad_norm": 0.0019585881382226944, | |
| "learning_rate": 1.89873417721519e-05, | |
| "loss": 0.0003, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 0.02020511031150818, | |
| "learning_rate": 1.8635724331926866e-05, | |
| "loss": 0.0084, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 0.002657042583450675, | |
| "learning_rate": 1.828410689170183e-05, | |
| "loss": 0.0003, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "grad_norm": 0.002752606989815831, | |
| "learning_rate": 1.7932489451476795e-05, | |
| "loss": 0.0278, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "grad_norm": 0.4717991352081299, | |
| "learning_rate": 1.758087201125176e-05, | |
| "loss": 0.0004, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 0.01203097589313984, | |
| "learning_rate": 1.7229254571026725e-05, | |
| "loss": 0.0271, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 0.001858622650615871, | |
| "learning_rate": 1.6877637130801688e-05, | |
| "loss": 0.0341, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 0.021344369277358055, | |
| "learning_rate": 1.6526019690576654e-05, | |
| "loss": 0.0198, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "grad_norm": 0.05389473959803581, | |
| "learning_rate": 1.617440225035162e-05, | |
| "loss": 0.0022, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "grad_norm": 0.0023719954770058393, | |
| "learning_rate": 1.5822784810126583e-05, | |
| "loss": 0.0069, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.0039686476811766624, | |
| "learning_rate": 1.547116736990155e-05, | |
| "loss": 0.0007, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 0.004471825901418924, | |
| "learning_rate": 1.5119549929676513e-05, | |
| "loss": 0.0003, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "grad_norm": 0.0025274273939430714, | |
| "learning_rate": 1.4767932489451477e-05, | |
| "loss": 0.0004, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 0.005175269674509764, | |
| "learning_rate": 1.4416315049226442e-05, | |
| "loss": 0.0037, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "grad_norm": 0.4415951073169708, | |
| "learning_rate": 1.4064697609001406e-05, | |
| "loss": 0.0022, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.02205347828567028, | |
| "learning_rate": 1.3713080168776371e-05, | |
| "loss": 0.0049, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "grad_norm": 0.0073618339374661446, | |
| "learning_rate": 1.3361462728551336e-05, | |
| "loss": 0.0005, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 0.01835988275706768, | |
| "learning_rate": 1.3009845288326302e-05, | |
| "loss": 0.0002, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "grad_norm": 0.006863409653306007, | |
| "learning_rate": 1.2658227848101267e-05, | |
| "loss": 0.0012, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "grad_norm": 0.009447610937058926, | |
| "learning_rate": 1.2306610407876231e-05, | |
| "loss": 0.0008, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 0.0040683564729988575, | |
| "learning_rate": 1.1954992967651196e-05, | |
| "loss": 0.0003, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "grad_norm": 0.0015517990104854107, | |
| "learning_rate": 1.160337552742616e-05, | |
| "loss": 0.0004, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "grad_norm": 0.0038061882369220257, | |
| "learning_rate": 1.1251758087201125e-05, | |
| "loss": 0.0002, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9992091735863978, | |
| "eval_loss": 0.0019393304828554392, | |
| "eval_runtime": 17.2093, | |
| "eval_samples_per_second": 146.955, | |
| "eval_steps_per_second": 18.42, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "grad_norm": 0.021453727036714554, | |
| "learning_rate": 1.0900140646976091e-05, | |
| "loss": 0.0005, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 0.003984913229942322, | |
| "learning_rate": 1.0548523206751056e-05, | |
| "loss": 0.0005, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 0.00517477560788393, | |
| "learning_rate": 1.019690576652602e-05, | |
| "loss": 0.001, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "grad_norm": 0.002068225760012865, | |
| "learning_rate": 9.845288326300985e-06, | |
| "loss": 0.0002, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "grad_norm": 0.00740728760138154, | |
| "learning_rate": 9.49367088607595e-06, | |
| "loss": 0.0002, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 0.0014793493319302797, | |
| "learning_rate": 9.142053445850915e-06, | |
| "loss": 0.0017, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "grad_norm": 0.0028481709305197, | |
| "learning_rate": 8.79043600562588e-06, | |
| "loss": 0.0011, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "grad_norm": 0.06655070930719376, | |
| "learning_rate": 8.438818565400844e-06, | |
| "loss": 0.0009, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "grad_norm": 0.012430869042873383, | |
| "learning_rate": 8.08720112517581e-06, | |
| "loss": 0.0014, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "grad_norm": 0.0027373009361326694, | |
| "learning_rate": 7.735583684950775e-06, | |
| "loss": 0.0067, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "grad_norm": 0.0010446230880916119, | |
| "learning_rate": 7.3839662447257386e-06, | |
| "loss": 0.0001, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "grad_norm": 0.0015856565441936255, | |
| "learning_rate": 7.032348804500703e-06, | |
| "loss": 0.0002, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 0.0012825512094423175, | |
| "learning_rate": 6.680731364275668e-06, | |
| "loss": 0.0003, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "grad_norm": 0.0009867001790553331, | |
| "learning_rate": 6.329113924050633e-06, | |
| "loss": 0.0002, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "grad_norm": 0.0017112062778323889, | |
| "learning_rate": 5.977496483825598e-06, | |
| "loss": 0.0003, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 0.0032221453730016947, | |
| "learning_rate": 5.6258790436005626e-06, | |
| "loss": 0.0002, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "grad_norm": 0.003395942272618413, | |
| "learning_rate": 5.274261603375528e-06, | |
| "loss": 0.0001, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 0.0007977127097547054, | |
| "learning_rate": 4.922644163150493e-06, | |
| "loss": 0.0132, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "grad_norm": 0.0035839290358126163, | |
| "learning_rate": 4.571026722925457e-06, | |
| "loss": 0.0001, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "grad_norm": 2.0375587940216064, | |
| "learning_rate": 4.219409282700422e-06, | |
| "loss": 0.0166, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "grad_norm": 0.005914665758609772, | |
| "learning_rate": 3.867791842475387e-06, | |
| "loss": 0.0001, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "grad_norm": 0.6946305632591248, | |
| "learning_rate": 3.5161744022503516e-06, | |
| "loss": 0.0011, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "grad_norm": 0.003537875832989812, | |
| "learning_rate": 3.1645569620253167e-06, | |
| "loss": 0.0129, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "grad_norm": 0.01320402417331934, | |
| "learning_rate": 2.8129395218002813e-06, | |
| "loss": 0.0092, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "grad_norm": 0.10570605099201202, | |
| "learning_rate": 2.4613220815752463e-06, | |
| "loss": 0.0008, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "grad_norm": 0.0033983599860221148, | |
| "learning_rate": 2.109704641350211e-06, | |
| "loss": 0.0002, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "grad_norm": 0.0025147399865090847, | |
| "learning_rate": 1.7580872011251758e-06, | |
| "loss": 0.0001, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "grad_norm": 0.004903899040073156, | |
| "learning_rate": 1.4064697609001406e-06, | |
| "loss": 0.016, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "grad_norm": 0.002249601762741804, | |
| "learning_rate": 1.0548523206751055e-06, | |
| "loss": 0.0024, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "grad_norm": 0.005820867605507374, | |
| "learning_rate": 7.032348804500703e-07, | |
| "loss": 0.0003, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 0.46763402223587036, | |
| "learning_rate": 3.5161744022503516e-07, | |
| "loss": 0.0006, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.628798246383667, | |
| "learning_rate": 0.0, | |
| "loss": 0.0019, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9992091735863978, | |
| "eval_loss": 0.0018690477591007948, | |
| "eval_runtime": 17.1455, | |
| "eval_samples_per_second": 147.502, | |
| "eval_steps_per_second": 18.489, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 1580, | |
| "total_flos": 3.915825219204563e+18, | |
| "train_loss": 0.0232889701421461, | |
| "train_runtime": 841.0378, | |
| "train_samples_per_second": 60.122, | |
| "train_steps_per_second": 1.879 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1580, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 3.915825219204563e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |