| { | |
| "best_metric": 0.5454545454545454, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-isic217/checkpoint-245", | |
| "epoch": 48.97959183673469, | |
| "eval_steps": 500, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.40816326530612246, | |
| "grad_norm": 20.10276222229004, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 2.1681, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8163265306122449, | |
| "grad_norm": 18.530778884887695, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 2.2404, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9795918367346939, | |
| "eval_accuracy": 0.13636363636363635, | |
| "eval_loss": 2.1526713371276855, | |
| "eval_runtime": 3.7709, | |
| "eval_samples_per_second": 5.834, | |
| "eval_steps_per_second": 2.917, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.2244897959183674, | |
| "grad_norm": 19.15616798400879, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.1402, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.6326530612244898, | |
| "grad_norm": 18.088478088378906, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.0749, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.13636363636363635, | |
| "eval_loss": 2.115939140319824, | |
| "eval_runtime": 4.2644, | |
| "eval_samples_per_second": 5.159, | |
| "eval_steps_per_second": 2.58, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.0408163265306123, | |
| "grad_norm": 28.553932189941406, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 2.0201, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.4489795918367347, | |
| "grad_norm": 21.90468406677246, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.8697, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 29.221223831176758, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 1.947, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.979591836734694, | |
| "eval_accuracy": 0.13636363636363635, | |
| "eval_loss": 2.172260284423828, | |
| "eval_runtime": 3.7285, | |
| "eval_samples_per_second": 5.9, | |
| "eval_steps_per_second": 2.95, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 3.2653061224489797, | |
| "grad_norm": 26.88852882385254, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.8989, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.673469387755102, | |
| "grad_norm": 29.26166343688965, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.732, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.09090909090909091, | |
| "eval_loss": 2.154468536376953, | |
| "eval_runtime": 4.2209, | |
| "eval_samples_per_second": 5.212, | |
| "eval_steps_per_second": 2.606, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 4.081632653061225, | |
| "grad_norm": 39.108612060546875, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.6138, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.489795918367347, | |
| "grad_norm": 23.49874496459961, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 1.4534, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.8979591836734695, | |
| "grad_norm": 26.28005599975586, | |
| "learning_rate": 5e-05, | |
| "loss": 1.446, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.979591836734694, | |
| "eval_accuracy": 0.18181818181818182, | |
| "eval_loss": 2.2917678356170654, | |
| "eval_runtime": 3.8303, | |
| "eval_samples_per_second": 5.744, | |
| "eval_steps_per_second": 2.872, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 5.3061224489795915, | |
| "grad_norm": 38.8511962890625, | |
| "learning_rate": 4.9537037037037035e-05, | |
| "loss": 1.2513, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 29.820043563842773, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 1.1175, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.3181818181818182, | |
| "eval_loss": 1.9290556907653809, | |
| "eval_runtime": 4.301, | |
| "eval_samples_per_second": 5.115, | |
| "eval_steps_per_second": 2.558, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 6.122448979591836, | |
| "grad_norm": 36.40614700317383, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 1.3955, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.530612244897959, | |
| "grad_norm": 31.91864776611328, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 1.0427, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 6.938775510204081, | |
| "grad_norm": 35.82953643798828, | |
| "learning_rate": 4.768518518518519e-05, | |
| "loss": 1.1069, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.979591836734694, | |
| "eval_accuracy": 0.36363636363636365, | |
| "eval_loss": 1.9551143646240234, | |
| "eval_runtime": 4.268, | |
| "eval_samples_per_second": 5.155, | |
| "eval_steps_per_second": 2.577, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 7.346938775510204, | |
| "grad_norm": 30.75468635559082, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.9718, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 7.755102040816326, | |
| "grad_norm": 33.49174880981445, | |
| "learning_rate": 4.675925925925926e-05, | |
| "loss": 0.7932, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 2.0534465312957764, | |
| "eval_runtime": 4.3072, | |
| "eval_samples_per_second": 5.108, | |
| "eval_steps_per_second": 2.554, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 8.16326530612245, | |
| "grad_norm": 45.86684799194336, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.0077, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 34.891082763671875, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.7286, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 8.979591836734693, | |
| "grad_norm": 30.408119201660156, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 0.5994, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.979591836734693, | |
| "eval_accuracy": 0.36363636363636365, | |
| "eval_loss": 1.8135310411453247, | |
| "eval_runtime": 3.8949, | |
| "eval_samples_per_second": 5.648, | |
| "eval_steps_per_second": 2.824, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.387755102040817, | |
| "grad_norm": 20.37044906616211, | |
| "learning_rate": 4.490740740740741e-05, | |
| "loss": 0.7078, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 9.795918367346939, | |
| "grad_norm": 14.344355583190918, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.4671, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5454545454545454, | |
| "eval_loss": 1.7822338342666626, | |
| "eval_runtime": 4.1682, | |
| "eval_samples_per_second": 5.278, | |
| "eval_steps_per_second": 2.639, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 10.204081632653061, | |
| "grad_norm": 39.557579040527344, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 0.5962, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 10.612244897959183, | |
| "grad_norm": 7.858915328979492, | |
| "learning_rate": 4.351851851851852e-05, | |
| "loss": 0.4612, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 10.979591836734693, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.2529866695404053, | |
| "eval_runtime": 3.7854, | |
| "eval_samples_per_second": 5.812, | |
| "eval_steps_per_second": 2.906, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 11.020408163265307, | |
| "grad_norm": 28.87532615661621, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 0.4548, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 40.782554626464844, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.5538, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 11.83673469387755, | |
| "grad_norm": 17.471019744873047, | |
| "learning_rate": 4.212962962962963e-05, | |
| "loss": 0.4016, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 1.8437938690185547, | |
| "eval_runtime": 4.198, | |
| "eval_samples_per_second": 5.241, | |
| "eval_steps_per_second": 2.62, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 12.244897959183673, | |
| "grad_norm": 38.50642395019531, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.4786, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 12.653061224489797, | |
| "grad_norm": 30.62510871887207, | |
| "learning_rate": 4.1203703703703705e-05, | |
| "loss": 0.3947, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 12.979591836734693, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.8609530925750732, | |
| "eval_runtime": 3.8332, | |
| "eval_samples_per_second": 5.739, | |
| "eval_steps_per_second": 2.87, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 13.061224489795919, | |
| "grad_norm": 1.0034515857696533, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.3527, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 13.46938775510204, | |
| "grad_norm": 11.846328735351562, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 0.3415, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 13.877551020408163, | |
| "grad_norm": 9.749055862426758, | |
| "learning_rate": 3.981481481481482e-05, | |
| "loss": 0.5033, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.45454545454545453, | |
| "eval_loss": 1.8614706993103027, | |
| "eval_runtime": 4.1423, | |
| "eval_samples_per_second": 5.311, | |
| "eval_steps_per_second": 2.656, | |
| "step": 343 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 74.4493637084961, | |
| "learning_rate": 3.935185185185186e-05, | |
| "loss": 0.2709, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 14.693877551020408, | |
| "grad_norm": 34.6863899230957, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2846, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 14.979591836734693, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.5478615760803223, | |
| "eval_runtime": 3.7136, | |
| "eval_samples_per_second": 5.924, | |
| "eval_steps_per_second": 2.962, | |
| "step": 367 | |
| }, | |
| { | |
| "epoch": 15.10204081632653, | |
| "grad_norm": 21.048105239868164, | |
| "learning_rate": 3.8425925925925924e-05, | |
| "loss": 0.361, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 15.510204081632653, | |
| "grad_norm": 27.45090103149414, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 0.1265, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 15.918367346938776, | |
| "grad_norm": 2.5599782466888428, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2828, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5454545454545454, | |
| "eval_loss": 1.6410338878631592, | |
| "eval_runtime": 4.2352, | |
| "eval_samples_per_second": 5.195, | |
| "eval_steps_per_second": 2.597, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 16.3265306122449, | |
| "grad_norm": 44.01292037963867, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.4138, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 16.73469387755102, | |
| "grad_norm": 49.22306823730469, | |
| "learning_rate": 3.6574074074074076e-05, | |
| "loss": 0.3426, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 16.979591836734695, | |
| "eval_accuracy": 0.36363636363636365, | |
| "eval_loss": 1.9146943092346191, | |
| "eval_runtime": 3.73, | |
| "eval_samples_per_second": 5.898, | |
| "eval_steps_per_second": 2.949, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 20.12368392944336, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.3658, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 17.551020408163264, | |
| "grad_norm": 5.72607946395874, | |
| "learning_rate": 3.564814814814815e-05, | |
| "loss": 0.0983, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 17.959183673469386, | |
| "grad_norm": 127.7773208618164, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.3108, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.5454545454545454, | |
| "eval_loss": 1.4793992042541504, | |
| "eval_runtime": 4.4219, | |
| "eval_samples_per_second": 4.975, | |
| "eval_steps_per_second": 2.488, | |
| "step": 441 | |
| }, | |
| { | |
| "epoch": 18.367346938775512, | |
| "grad_norm": 2.2349815368652344, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.08, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 18.775510204081634, | |
| "grad_norm": 84.97185516357422, | |
| "learning_rate": 3.425925925925926e-05, | |
| "loss": 0.2129, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 18.979591836734695, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 1.776501178741455, | |
| "eval_runtime": 3.7003, | |
| "eval_samples_per_second": 5.946, | |
| "eval_steps_per_second": 2.973, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 19.183673469387756, | |
| "grad_norm": 6.593123912811279, | |
| "learning_rate": 3.3796296296296295e-05, | |
| "loss": 0.1714, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 19.591836734693878, | |
| "grad_norm": 2.1411988735198975, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1612, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.131197929382324, | |
| "learning_rate": 3.2870370370370375e-05, | |
| "loss": 0.1946, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.45454545454545453, | |
| "eval_loss": 2.16646671295166, | |
| "eval_runtime": 4.4203, | |
| "eval_samples_per_second": 4.977, | |
| "eval_steps_per_second": 2.489, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 20.408163265306122, | |
| "grad_norm": 4.898344039916992, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.2809, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 20.816326530612244, | |
| "grad_norm": 0.8991552591323853, | |
| "learning_rate": 3.194444444444444e-05, | |
| "loss": 0.1255, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 20.979591836734695, | |
| "eval_accuracy": 0.5454545454545454, | |
| "eval_loss": 1.8521960973739624, | |
| "eval_runtime": 4.0094, | |
| "eval_samples_per_second": 5.487, | |
| "eval_steps_per_second": 2.744, | |
| "step": 514 | |
| }, | |
| { | |
| "epoch": 21.224489795918366, | |
| "grad_norm": 0.9338099956512451, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.0503, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 21.632653061224488, | |
| "grad_norm": 1.708648920059204, | |
| "learning_rate": 3.101851851851852e-05, | |
| "loss": 0.2301, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.45454545454545453, | |
| "eval_loss": 2.028917074203491, | |
| "eval_runtime": 4.3302, | |
| "eval_samples_per_second": 5.081, | |
| "eval_steps_per_second": 2.54, | |
| "step": 539 | |
| }, | |
| { | |
| "epoch": 22.040816326530614, | |
| "grad_norm": 68.97888946533203, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1664, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 22.448979591836736, | |
| "grad_norm": 8.795487403869629, | |
| "learning_rate": 3.0092592592592593e-05, | |
| "loss": 0.0495, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 1.032072901725769, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.0909, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 22.979591836734695, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 2.051244020462036, | |
| "eval_runtime": 4.0425, | |
| "eval_samples_per_second": 5.442, | |
| "eval_steps_per_second": 2.721, | |
| "step": 563 | |
| }, | |
| { | |
| "epoch": 23.26530612244898, | |
| "grad_norm": 28.04718589782715, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.1032, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 23.6734693877551, | |
| "grad_norm": 24.228317260742188, | |
| "learning_rate": 2.8703703703703706e-05, | |
| "loss": 0.1724, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 2.3410568237304688, | |
| "eval_runtime": 4.2445, | |
| "eval_samples_per_second": 5.183, | |
| "eval_steps_per_second": 2.592, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 24.081632653061224, | |
| "grad_norm": 51.32829284667969, | |
| "learning_rate": 2.824074074074074e-05, | |
| "loss": 0.1376, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 24.489795918367346, | |
| "grad_norm": 8.50900936126709, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0941, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 24.897959183673468, | |
| "grad_norm": 53.41023254394531, | |
| "learning_rate": 2.7314814814814816e-05, | |
| "loss": 0.2256, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 24.979591836734695, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.162550687789917, | |
| "eval_runtime": 3.7513, | |
| "eval_samples_per_second": 5.865, | |
| "eval_steps_per_second": 2.932, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 25.306122448979593, | |
| "grad_norm": 20.35555076599121, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 0.2228, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "grad_norm": 5.79302978515625, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 0.2471, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 2.0551793575286865, | |
| "eval_runtime": 4.2279, | |
| "eval_samples_per_second": 5.204, | |
| "eval_steps_per_second": 2.602, | |
| "step": 637 | |
| }, | |
| { | |
| "epoch": 26.122448979591837, | |
| "grad_norm": 57.79982376098633, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.1514, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 26.53061224489796, | |
| "grad_norm": 9.047138214111328, | |
| "learning_rate": 2.5462962962962965e-05, | |
| "loss": 0.056, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 26.93877551020408, | |
| "grad_norm": 8.557161331176758, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0671, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 26.979591836734695, | |
| "eval_accuracy": 0.5454545454545454, | |
| "eval_loss": 1.9339114427566528, | |
| "eval_runtime": 3.8162, | |
| "eval_samples_per_second": 5.765, | |
| "eval_steps_per_second": 2.882, | |
| "step": 661 | |
| }, | |
| { | |
| "epoch": 27.346938775510203, | |
| "grad_norm": 0.14792662858963013, | |
| "learning_rate": 2.4537037037037038e-05, | |
| "loss": 0.1409, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 27.755102040816325, | |
| "grad_norm": 112.82161712646484, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.2563, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.45454545454545453, | |
| "eval_loss": 2.2506721019744873, | |
| "eval_runtime": 4.1968, | |
| "eval_samples_per_second": 5.242, | |
| "eval_steps_per_second": 2.621, | |
| "step": 686 | |
| }, | |
| { | |
| "epoch": 28.163265306122447, | |
| "grad_norm": 0.717983067035675, | |
| "learning_rate": 2.361111111111111e-05, | |
| "loss": 0.0499, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 47.17780303955078, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.1607, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 28.979591836734695, | |
| "grad_norm": 102.77259063720703, | |
| "learning_rate": 2.2685185185185187e-05, | |
| "loss": 0.1865, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 28.979591836734695, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.0703585147857666, | |
| "eval_runtime": 3.7043, | |
| "eval_samples_per_second": 5.939, | |
| "eval_steps_per_second": 2.969, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 29.387755102040817, | |
| "grad_norm": 0.42554885149002075, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1905, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 29.79591836734694, | |
| "grad_norm": 92.93157196044922, | |
| "learning_rate": 2.175925925925926e-05, | |
| "loss": 0.0477, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.3181818181818182, | |
| "eval_loss": 2.839505195617676, | |
| "eval_runtime": 4.1643, | |
| "eval_samples_per_second": 5.283, | |
| "eval_steps_per_second": 2.642, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 30.20408163265306, | |
| "grad_norm": 10.255499839782715, | |
| "learning_rate": 2.1296296296296296e-05, | |
| "loss": 0.2298, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 30.612244897959183, | |
| "grad_norm": 32.9160041809082, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.0931, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 30.979591836734695, | |
| "eval_accuracy": 0.36363636363636365, | |
| "eval_loss": 2.9483964443206787, | |
| "eval_runtime": 3.6962, | |
| "eval_samples_per_second": 5.952, | |
| "eval_steps_per_second": 2.976, | |
| "step": 759 | |
| }, | |
| { | |
| "epoch": 31.020408163265305, | |
| "grad_norm": 3.2979772090911865, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.0643, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 31.428571428571427, | |
| "grad_norm": 1.9711153507232666, | |
| "learning_rate": 1.990740740740741e-05, | |
| "loss": 0.1815, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 31.836734693877553, | |
| "grad_norm": 39.575653076171875, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.047, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.45454545454545453, | |
| "eval_loss": 2.548579216003418, | |
| "eval_runtime": 4.2179, | |
| "eval_samples_per_second": 5.216, | |
| "eval_steps_per_second": 2.608, | |
| "step": 784 | |
| }, | |
| { | |
| "epoch": 32.244897959183675, | |
| "grad_norm": 0.1791549175977707, | |
| "learning_rate": 1.8981481481481482e-05, | |
| "loss": 0.1034, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 32.6530612244898, | |
| "grad_norm": 18.35671043395996, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.165, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 32.97959183673469, | |
| "eval_accuracy": 0.45454545454545453, | |
| "eval_loss": 2.601107120513916, | |
| "eval_runtime": 3.894, | |
| "eval_samples_per_second": 5.65, | |
| "eval_steps_per_second": 2.825, | |
| "step": 808 | |
| }, | |
| { | |
| "epoch": 33.06122448979592, | |
| "grad_norm": 88.95061492919922, | |
| "learning_rate": 1.8055555555555555e-05, | |
| "loss": 0.0639, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 33.46938775510204, | |
| "grad_norm": 0.11740878969430923, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 0.032, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 33.87755102040816, | |
| "grad_norm": 8.803248405456543, | |
| "learning_rate": 1.712962962962963e-05, | |
| "loss": 0.0203, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.3598248958587646, | |
| "eval_runtime": 4.1023, | |
| "eval_samples_per_second": 5.363, | |
| "eval_steps_per_second": 2.681, | |
| "step": 833 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 3.050877571105957, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0173, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 34.69387755102041, | |
| "grad_norm": 10.551043510437012, | |
| "learning_rate": 1.6203703703703704e-05, | |
| "loss": 0.0143, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 34.97959183673469, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 2.589207649230957, | |
| "eval_runtime": 4.2169, | |
| "eval_samples_per_second": 5.217, | |
| "eval_steps_per_second": 2.609, | |
| "step": 857 | |
| }, | |
| { | |
| "epoch": 35.10204081632653, | |
| "grad_norm": 2.492408514022827, | |
| "learning_rate": 1.574074074074074e-05, | |
| "loss": 0.1669, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 35.51020408163265, | |
| "grad_norm": 28.920665740966797, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 0.0546, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 35.91836734693877, | |
| "grad_norm": 0.1343473196029663, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.0248, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 2.8362252712249756, | |
| "eval_runtime": 4.252, | |
| "eval_samples_per_second": 5.174, | |
| "eval_steps_per_second": 2.587, | |
| "step": 882 | |
| }, | |
| { | |
| "epoch": 36.326530612244895, | |
| "grad_norm": 0.00863693282008171, | |
| "learning_rate": 1.4351851851851853e-05, | |
| "loss": 0.0432, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 36.734693877551024, | |
| "grad_norm": 1.786009430885315, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0812, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 36.97959183673469, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 2.4658091068267822, | |
| "eval_runtime": 4.0409, | |
| "eval_samples_per_second": 5.444, | |
| "eval_steps_per_second": 2.722, | |
| "step": 906 | |
| }, | |
| { | |
| "epoch": 37.142857142857146, | |
| "grad_norm": 5.971357822418213, | |
| "learning_rate": 1.3425925925925928e-05, | |
| "loss": 0.1312, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 37.55102040816327, | |
| "grad_norm": 9.33790111541748, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.0453, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 37.95918367346939, | |
| "grad_norm": 13.613574028015137, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0662, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 2.640266180038452, | |
| "eval_runtime": 4.0647, | |
| "eval_samples_per_second": 5.412, | |
| "eval_steps_per_second": 2.706, | |
| "step": 931 | |
| }, | |
| { | |
| "epoch": 38.36734693877551, | |
| "grad_norm": 17.910255432128906, | |
| "learning_rate": 1.2037037037037037e-05, | |
| "loss": 0.1061, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 38.775510204081634, | |
| "grad_norm": 5.941530227661133, | |
| "learning_rate": 1.1574074074074075e-05, | |
| "loss": 0.1855, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 38.97959183673469, | |
| "eval_accuracy": 0.45454545454545453, | |
| "eval_loss": 2.604184150695801, | |
| "eval_runtime": 3.8038, | |
| "eval_samples_per_second": 5.784, | |
| "eval_steps_per_second": 2.892, | |
| "step": 955 | |
| }, | |
| { | |
| "epoch": 39.183673469387756, | |
| "grad_norm": 0.057647667825222015, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0598, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 39.59183673469388, | |
| "grad_norm": 1.981195092201233, | |
| "learning_rate": 1.0648148148148148e-05, | |
| "loss": 0.0107, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 74.54513549804688, | |
| "learning_rate": 1.0185185185185185e-05, | |
| "loss": 0.03, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.5595407485961914, | |
| "eval_runtime": 4.2131, | |
| "eval_samples_per_second": 5.222, | |
| "eval_steps_per_second": 2.611, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 40.40816326530612, | |
| "grad_norm": 3.7874374389648438, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.0882, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 40.816326530612244, | |
| "grad_norm": 0.011240988969802856, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.1117, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 40.97959183673469, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.4715521335601807, | |
| "eval_runtime": 3.7805, | |
| "eval_samples_per_second": 5.819, | |
| "eval_steps_per_second": 2.91, | |
| "step": 1004 | |
| }, | |
| { | |
| "epoch": 41.224489795918366, | |
| "grad_norm": 30.458051681518555, | |
| "learning_rate": 8.796296296296297e-06, | |
| "loss": 0.0819, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 41.63265306122449, | |
| "grad_norm": 0.2596540153026581, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0466, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.325749397277832, | |
| "eval_runtime": 4.0793, | |
| "eval_samples_per_second": 5.393, | |
| "eval_steps_per_second": 2.697, | |
| "step": 1029 | |
| }, | |
| { | |
| "epoch": 42.04081632653061, | |
| "grad_norm": 0.8128257393836975, | |
| "learning_rate": 7.87037037037037e-06, | |
| "loss": 0.2931, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 42.44897959183673, | |
| "grad_norm": 1.8079004287719727, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.0667, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "grad_norm": 13.098880767822266, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.1349, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 42.97959183673469, | |
| "eval_accuracy": 0.45454545454545453, | |
| "eval_loss": 2.5544846057891846, | |
| "eval_runtime": 3.7821, | |
| "eval_samples_per_second": 5.817, | |
| "eval_steps_per_second": 2.908, | |
| "step": 1053 | |
| }, | |
| { | |
| "epoch": 43.265306122448976, | |
| "grad_norm": 0.03145942464470863, | |
| "learning_rate": 6.481481481481481e-06, | |
| "loss": 0.0184, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 43.673469387755105, | |
| "grad_norm": 0.008914113976061344, | |
| "learning_rate": 6.0185185185185185e-06, | |
| "loss": 0.0069, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.5814802646636963, | |
| "eval_runtime": 4.2458, | |
| "eval_samples_per_second": 5.182, | |
| "eval_steps_per_second": 2.591, | |
| "step": 1078 | |
| }, | |
| { | |
| "epoch": 44.08163265306123, | |
| "grad_norm": 0.0782785564661026, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1017, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 44.48979591836735, | |
| "grad_norm": 0.24598410725593567, | |
| "learning_rate": 5.092592592592592e-06, | |
| "loss": 0.0344, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 44.89795918367347, | |
| "grad_norm": 0.017518645152449608, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.0468, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 44.97959183673469, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.3666675090789795, | |
| "eval_runtime": 3.9709, | |
| "eval_samples_per_second": 5.54, | |
| "eval_steps_per_second": 2.77, | |
| "step": 1102 | |
| }, | |
| { | |
| "epoch": 45.30612244897959, | |
| "grad_norm": 3.6144909858703613, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.0372, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 0.018579425290226936, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.1807, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.472862482070923, | |
| "eval_runtime": 4.5348, | |
| "eval_samples_per_second": 4.851, | |
| "eval_steps_per_second": 2.426, | |
| "step": 1127 | |
| }, | |
| { | |
| "epoch": 46.12244897959184, | |
| "grad_norm": 6.9941864013671875, | |
| "learning_rate": 3.2407407407407406e-06, | |
| "loss": 0.0091, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 46.53061224489796, | |
| "grad_norm": 1.61014986038208, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0388, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 46.93877551020408, | |
| "grad_norm": 0.021990543231368065, | |
| "learning_rate": 2.3148148148148148e-06, | |
| "loss": 0.0667, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 46.97959183673469, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.496938467025757, | |
| "eval_runtime": 3.9012, | |
| "eval_samples_per_second": 5.639, | |
| "eval_steps_per_second": 2.82, | |
| "step": 1151 | |
| }, | |
| { | |
| "epoch": 47.3469387755102, | |
| "grad_norm": 0.5512164235115051, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.0191, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 47.755102040816325, | |
| "grad_norm": 0.2976542115211487, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 0.0199, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.5520730018615723, | |
| "eval_runtime": 4.4614, | |
| "eval_samples_per_second": 4.931, | |
| "eval_steps_per_second": 2.466, | |
| "step": 1176 | |
| }, | |
| { | |
| "epoch": 48.16326530612245, | |
| "grad_norm": 0.37567588686943054, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 0.0806, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "grad_norm": 2.0820977687835693, | |
| "learning_rate": 4.6296296296296297e-07, | |
| "loss": 0.1011, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 48.97959183673469, | |
| "grad_norm": 0.018370352685451508, | |
| "learning_rate": 0.0, | |
| "loss": 0.2716, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 48.97959183673469, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 2.5371408462524414, | |
| "eval_runtime": 3.8373, | |
| "eval_samples_per_second": 5.733, | |
| "eval_steps_per_second": 2.867, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 48.97959183673469, | |
| "step": 1200, | |
| "total_flos": 2.374708462608384e+17, | |
| "train_loss": 0.41171714147552846, | |
| "train_runtime": 1608.8411, | |
| "train_samples_per_second": 6.06, | |
| "train_steps_per_second": 0.746 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.374708462608384e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |