| { | |
| "best_metric": 0.008771373890340328, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-Kontur-competition/checkpoint-807", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1345, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 5.5162248611450195, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.6819, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 3.1403090953826904, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.5394, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 2.9430086612701416, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.3795, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 3.1442313194274902, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.2726, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 2.8489677906036377, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1909, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 2.6733438968658447, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0903, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 6.4875969886779785, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.064, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 10.464859008789062, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.0439, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 3.629943609237671, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0384, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 12.326812744140625, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.0418, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 4.8540754318237305, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.0378, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 7.512868881225586, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0261, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.206182479858398, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.0317, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 4.706695556640625, | |
| "learning_rate": 4.979338842975207e-05, | |
| "loss": 0.0353, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1.3356839418411255, | |
| "learning_rate": 4.9380165289256205e-05, | |
| "loss": 0.0261, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.30266404151916504, | |
| "learning_rate": 4.896694214876033e-05, | |
| "loss": 0.0174, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 4.922573566436768, | |
| "learning_rate": 4.855371900826447e-05, | |
| "loss": 0.0533, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 2.4459125995635986, | |
| "learning_rate": 4.8140495867768596e-05, | |
| "loss": 0.0275, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 4.6485915184021, | |
| "learning_rate": 4.772727272727273e-05, | |
| "loss": 0.0382, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 7.907899379730225, | |
| "learning_rate": 4.731404958677686e-05, | |
| "loss": 0.0312, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.321392297744751, | |
| "learning_rate": 4.6900826446280993e-05, | |
| "loss": 0.0174, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.5613566040992737, | |
| "learning_rate": 4.648760330578513e-05, | |
| "loss": 0.0128, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.357917845249176, | |
| "learning_rate": 4.607438016528926e-05, | |
| "loss": 0.0173, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 1.047433853149414, | |
| "learning_rate": 4.566115702479339e-05, | |
| "loss": 0.0231, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.5947940349578857, | |
| "learning_rate": 4.524793388429752e-05, | |
| "loss": 0.0279, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 1.0275371074676514, | |
| "learning_rate": 4.4834710743801654e-05, | |
| "loss": 0.0216, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.06352325528860092, | |
| "eval_runtime": 43.5548, | |
| "eval_samples_per_second": 87.82, | |
| "eval_steps_per_second": 2.755, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 8.136920928955078, | |
| "learning_rate": 4.442148760330579e-05, | |
| "loss": 0.0226, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 5.6582255363464355, | |
| "learning_rate": 4.400826446280992e-05, | |
| "loss": 0.0142, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 7.312340259552002, | |
| "learning_rate": 4.359504132231405e-05, | |
| "loss": 0.0298, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.31576940417289734, | |
| "learning_rate": 4.318181818181819e-05, | |
| "loss": 0.018, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 14.974953651428223, | |
| "learning_rate": 4.2768595041322315e-05, | |
| "loss": 0.0367, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 4.256837844848633, | |
| "learning_rate": 4.235537190082644e-05, | |
| "loss": 0.0064, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 0.972306489944458, | |
| "learning_rate": 4.194214876033058e-05, | |
| "loss": 0.014, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 0.03949430584907532, | |
| "learning_rate": 4.152892561983471e-05, | |
| "loss": 0.0121, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.8921181559562683, | |
| "learning_rate": 4.111570247933885e-05, | |
| "loss": 0.0135, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 4.770934581756592, | |
| "learning_rate": 4.0702479338842975e-05, | |
| "loss": 0.0279, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 12.435590744018555, | |
| "learning_rate": 4.028925619834711e-05, | |
| "loss": 0.0263, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 4.150902271270752, | |
| "learning_rate": 3.9876033057851245e-05, | |
| "loss": 0.0169, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 1.9341884851455688, | |
| "learning_rate": 3.946280991735537e-05, | |
| "loss": 0.0126, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 2.019587516784668, | |
| "learning_rate": 3.90495867768595e-05, | |
| "loss": 0.0075, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.34746670722961426, | |
| "learning_rate": 3.8636363636363636e-05, | |
| "loss": 0.0186, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 4.3460187911987305, | |
| "learning_rate": 3.822314049586777e-05, | |
| "loss": 0.015, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.06722448021173477, | |
| "learning_rate": 3.7809917355371906e-05, | |
| "loss": 0.0314, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 5.969712257385254, | |
| "learning_rate": 3.7396694214876034e-05, | |
| "loss": 0.0132, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 5.939075946807861, | |
| "learning_rate": 3.698347107438017e-05, | |
| "loss": 0.0191, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 2.4963324069976807, | |
| "learning_rate": 3.65702479338843e-05, | |
| "loss": 0.0107, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 4.111177921295166, | |
| "learning_rate": 3.615702479338843e-05, | |
| "loss": 0.0204, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 0.07025858759880066, | |
| "learning_rate": 3.574380165289256e-05, | |
| "loss": 0.0142, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 9.363666534423828, | |
| "learning_rate": 3.5330578512396694e-05, | |
| "loss": 0.0251, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 2.941878080368042, | |
| "learning_rate": 3.491735537190083e-05, | |
| "loss": 0.0123, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 5.7476348876953125, | |
| "learning_rate": 3.4504132231404964e-05, | |
| "loss": 0.0096, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 27.465803146362305, | |
| "learning_rate": 3.409090909090909e-05, | |
| "loss": 0.028, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "grad_norm": 0.047597743570804596, | |
| "learning_rate": 3.367768595041322e-05, | |
| "loss": 0.0286, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.0467231348156929, | |
| "eval_runtime": 43.1197, | |
| "eval_samples_per_second": 88.707, | |
| "eval_steps_per_second": 2.783, | |
| "step": 538 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 0.10430701822042465, | |
| "learning_rate": 3.3264462809917355e-05, | |
| "loss": 0.009, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 4.288072109222412, | |
| "learning_rate": 3.285123966942149e-05, | |
| "loss": 0.0041, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.3560419082641602, | |
| "learning_rate": 3.243801652892562e-05, | |
| "loss": 0.0175, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 5.84924840927124, | |
| "learning_rate": 3.202479338842975e-05, | |
| "loss": 0.0095, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 8.711688041687012, | |
| "learning_rate": 3.161157024793389e-05, | |
| "loss": 0.028, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 1.378418207168579, | |
| "learning_rate": 3.119834710743802e-05, | |
| "loss": 0.0244, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 4.585868835449219, | |
| "learning_rate": 3.078512396694215e-05, | |
| "loss": 0.0045, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 14.25820541381836, | |
| "learning_rate": 3.0371900826446282e-05, | |
| "loss": 0.0217, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.5124061107635498, | |
| "learning_rate": 2.9958677685950414e-05, | |
| "loss": 0.0015, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 5.597954750061035, | |
| "learning_rate": 2.954545454545455e-05, | |
| "loss": 0.0096, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 4.180160045623779, | |
| "learning_rate": 2.9132231404958676e-05, | |
| "loss": 0.016, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 1.3298313617706299, | |
| "learning_rate": 2.871900826446281e-05, | |
| "loss": 0.0104, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 0.11605699360370636, | |
| "learning_rate": 2.8305785123966943e-05, | |
| "loss": 0.0176, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 0.0669424757361412, | |
| "learning_rate": 2.7892561983471078e-05, | |
| "loss": 0.0043, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 0.7822141647338867, | |
| "learning_rate": 2.7479338842975206e-05, | |
| "loss": 0.0041, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 0.9650640487670898, | |
| "learning_rate": 2.7066115702479337e-05, | |
| "loss": 0.0096, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 10.423120498657227, | |
| "learning_rate": 2.6652892561983472e-05, | |
| "loss": 0.0065, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 10.540702819824219, | |
| "learning_rate": 2.6239669421487607e-05, | |
| "loss": 0.0068, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 0.06544263660907745, | |
| "learning_rate": 2.5826446280991735e-05, | |
| "loss": 0.0142, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "grad_norm": 1.903037428855896, | |
| "learning_rate": 2.5413223140495866e-05, | |
| "loss": 0.0079, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 8.562331199645996, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0109, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 2.383655548095703, | |
| "learning_rate": 2.4586776859504136e-05, | |
| "loss": 0.0063, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 6.162371635437012, | |
| "learning_rate": 2.4173553719008264e-05, | |
| "loss": 0.0118, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 0.7068905234336853, | |
| "learning_rate": 2.37603305785124e-05, | |
| "loss": 0.016, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 0.07612281292676926, | |
| "learning_rate": 2.334710743801653e-05, | |
| "loss": 0.008, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 5.8349199295043945, | |
| "learning_rate": 2.2933884297520665e-05, | |
| "loss": 0.0091, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 0.30163004994392395, | |
| "learning_rate": 2.2520661157024793e-05, | |
| "loss": 0.0252, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.008771373890340328, | |
| "eval_runtime": 43.1769, | |
| "eval_samples_per_second": 88.589, | |
| "eval_steps_per_second": 2.779, | |
| "step": 807 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 1.3629869222640991, | |
| "learning_rate": 2.2107438016528928e-05, | |
| "loss": 0.0095, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 1.119672179222107, | |
| "learning_rate": 2.169421487603306e-05, | |
| "loss": 0.006, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 8.719965934753418, | |
| "learning_rate": 2.128099173553719e-05, | |
| "loss": 0.0144, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.12227091938257217, | |
| "learning_rate": 2.0867768595041323e-05, | |
| "loss": 0.0033, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 1.8717501163482666, | |
| "learning_rate": 2.0454545454545457e-05, | |
| "loss": 0.0027, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 6.553406238555908, | |
| "learning_rate": 2.004132231404959e-05, | |
| "loss": 0.0133, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 2.087083101272583, | |
| "learning_rate": 1.962809917355372e-05, | |
| "loss": 0.0134, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "grad_norm": 1.4037662744522095, | |
| "learning_rate": 1.9214876033057852e-05, | |
| "loss": 0.0031, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "grad_norm": 2.297409772872925, | |
| "learning_rate": 1.8801652892561987e-05, | |
| "loss": 0.0162, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 0.025398077443242073, | |
| "learning_rate": 1.8388429752066118e-05, | |
| "loss": 0.0061, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "grad_norm": 0.037510234862565994, | |
| "learning_rate": 1.797520661157025e-05, | |
| "loss": 0.0064, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "grad_norm": 0.3090139627456665, | |
| "learning_rate": 1.756198347107438e-05, | |
| "loss": 0.004, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "grad_norm": 3.6702754497528076, | |
| "learning_rate": 1.7148760330578516e-05, | |
| "loss": 0.0046, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "grad_norm": 0.22104105353355408, | |
| "learning_rate": 1.6735537190082644e-05, | |
| "loss": 0.0003, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "grad_norm": 2.2657175064086914, | |
| "learning_rate": 1.632231404958678e-05, | |
| "loss": 0.0048, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "grad_norm": 0.0027564549818634987, | |
| "learning_rate": 1.590909090909091e-05, | |
| "loss": 0.0112, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "grad_norm": 0.024479985237121582, | |
| "learning_rate": 1.549586776859504e-05, | |
| "loss": 0.0031, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 0.3030235767364502, | |
| "learning_rate": 1.5082644628099175e-05, | |
| "loss": 0.0031, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 0.01274044532328844, | |
| "learning_rate": 1.4669421487603308e-05, | |
| "loss": 0.0086, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "grad_norm": 0.9196871519088745, | |
| "learning_rate": 1.4256198347107438e-05, | |
| "loss": 0.01, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.09063485264778137, | |
| "learning_rate": 1.3842975206611573e-05, | |
| "loss": 0.0043, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "grad_norm": 0.008292087353765965, | |
| "learning_rate": 1.3429752066115702e-05, | |
| "loss": 0.0028, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 0.17935311794281006, | |
| "learning_rate": 1.3016528925619837e-05, | |
| "loss": 0.006, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 0.2591820955276489, | |
| "learning_rate": 1.2603305785123967e-05, | |
| "loss": 0.0072, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 0.8280309438705444, | |
| "learning_rate": 1.21900826446281e-05, | |
| "loss": 0.0044, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "grad_norm": 1.8738752603530884, | |
| "learning_rate": 1.1776859504132231e-05, | |
| "loss": 0.0033, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "grad_norm": 0.01168906595557928, | |
| "learning_rate": 1.1363636363636365e-05, | |
| "loss": 0.0003, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.033934369683265686, | |
| "eval_runtime": 42.9173, | |
| "eval_samples_per_second": 89.125, | |
| "eval_steps_per_second": 2.796, | |
| "step": 1076 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 1.7786719799041748, | |
| "learning_rate": 1.0950413223140496e-05, | |
| "loss": 0.005, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 0.44828712940216064, | |
| "learning_rate": 1.0537190082644628e-05, | |
| "loss": 0.001, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 0.0037061686161905527, | |
| "learning_rate": 1.012396694214876e-05, | |
| "loss": 0.0005, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "grad_norm": 7.8463358879089355, | |
| "learning_rate": 9.710743801652892e-06, | |
| "loss": 0.0054, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 0.07012557983398438, | |
| "learning_rate": 9.297520661157025e-06, | |
| "loss": 0.0037, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 2.830775260925293, | |
| "learning_rate": 8.884297520661157e-06, | |
| "loss": 0.0015, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "grad_norm": 0.2861256003379822, | |
| "learning_rate": 8.47107438016529e-06, | |
| "loss": 0.0042, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "grad_norm": 0.9108858108520508, | |
| "learning_rate": 8.057851239669421e-06, | |
| "loss": 0.0003, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "grad_norm": 0.06316760182380676, | |
| "learning_rate": 7.644628099173553e-06, | |
| "loss": 0.0002, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "grad_norm": 0.006003058515489101, | |
| "learning_rate": 7.231404958677686e-06, | |
| "loss": 0.001, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "grad_norm": 0.0066665462218225, | |
| "learning_rate": 6.818181818181818e-06, | |
| "loss": 0.012, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "grad_norm": 7.692056179046631, | |
| "learning_rate": 6.404958677685951e-06, | |
| "loss": 0.0029, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "grad_norm": 0.01048464234918356, | |
| "learning_rate": 5.991735537190083e-06, | |
| "loss": 0.0007, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 0.016681592911481857, | |
| "learning_rate": 5.578512396694215e-06, | |
| "loss": 0.0015, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "grad_norm": 0.035201072692871094, | |
| "learning_rate": 5.1652892561983475e-06, | |
| "loss": 0.0028, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "grad_norm": 0.05388149991631508, | |
| "learning_rate": 4.75206611570248e-06, | |
| "loss": 0.0008, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "grad_norm": 0.14326803386211395, | |
| "learning_rate": 4.338842975206612e-06, | |
| "loss": 0.0001, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "grad_norm": 0.001435384969227016, | |
| "learning_rate": 3.925619834710744e-06, | |
| "loss": 0.0048, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "grad_norm": 0.11075280606746674, | |
| "learning_rate": 3.5123966942148763e-06, | |
| "loss": 0.0006, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 0.05017843842506409, | |
| "learning_rate": 3.0991735537190086e-06, | |
| "loss": 0.0001, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "grad_norm": 0.002463772427290678, | |
| "learning_rate": 2.685950413223141e-06, | |
| "loss": 0.0003, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 13.101515769958496, | |
| "learning_rate": 2.2727272727272728e-06, | |
| "loss": 0.0011, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "grad_norm": 0.003497660392895341, | |
| "learning_rate": 1.859504132231405e-06, | |
| "loss": 0.0014, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "grad_norm": 0.07368556410074234, | |
| "learning_rate": 1.4462809917355374e-06, | |
| "loss": 0.0013, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "grad_norm": 0.0647512823343277, | |
| "learning_rate": 1.0330578512396695e-06, | |
| "loss": 0.0045, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "grad_norm": 0.0005362280062399805, | |
| "learning_rate": 6.198347107438017e-07, | |
| "loss": 0.0141, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "grad_norm": 0.001157268532551825, | |
| "learning_rate": 2.066115702479339e-07, | |
| "loss": 0.0019, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.012286508455872536, | |
| "eval_runtime": 42.4964, | |
| "eval_samples_per_second": 90.008, | |
| "eval_steps_per_second": 2.824, | |
| "step": 1345 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 1345, | |
| "total_flos": 4.2775914245208883e+18, | |
| "train_loss": 0.028647772959333094, | |
| "train_runtime": 3141.2683, | |
| "train_samples_per_second": 54.785, | |
| "train_steps_per_second": 0.428 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1345, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 4.2775914245208883e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |