| { | |
| "best_metric": 0.9713785046728972, | |
| "best_model_checkpoint": "swin-large-patch4-window7-224-in22k-finetuned-lora-medmnistv2/checkpoint-1870", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 1870, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 2.163198709487915, | |
| "learning_rate": 0.004973262032085562, | |
| "loss": 1.5101, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 1.4286870956420898, | |
| "learning_rate": 0.004946524064171123, | |
| "loss": 0.8667, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.847931981086731, | |
| "learning_rate": 0.004919786096256685, | |
| "loss": 0.7414, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 1.5748757123947144, | |
| "learning_rate": 0.004893048128342246, | |
| "loss": 0.755, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 2.017432928085327, | |
| "learning_rate": 0.004866310160427808, | |
| "loss": 0.6683, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.5988194942474365, | |
| "learning_rate": 0.004839572192513369, | |
| "loss": 0.7084, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.7127466201782227, | |
| "learning_rate": 0.004812834224598931, | |
| "loss": 0.6459, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 2.1388797760009766, | |
| "learning_rate": 0.004786096256684492, | |
| "loss": 0.7116, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 2.5939793586730957, | |
| "learning_rate": 0.004759358288770054, | |
| "loss": 0.5753, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 1.463460087776184, | |
| "learning_rate": 0.004732620320855615, | |
| "loss": 0.5938, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 1.9902774095535278, | |
| "learning_rate": 0.004705882352941177, | |
| "loss": 0.5525, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 1.881441593170166, | |
| "learning_rate": 0.004679144385026738, | |
| "loss": 0.5788, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.161348581314087, | |
| "learning_rate": 0.0046524064171123, | |
| "loss": 0.5378, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 1.5160846710205078, | |
| "learning_rate": 0.0046256684491978615, | |
| "loss": 0.479, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.4215080738067627, | |
| "learning_rate": 0.004598930481283423, | |
| "loss": 0.5123, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 1.2568920850753784, | |
| "learning_rate": 0.004572192513368984, | |
| "loss": 0.5499, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.9570059180259705, | |
| "learning_rate": 0.004545454545454545, | |
| "loss": 0.4845, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.3021810054779053, | |
| "learning_rate": 0.004518716577540107, | |
| "loss": 0.5141, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9065420560747663, | |
| "eval_f1": 0.8872552707825272, | |
| "eval_loss": 0.2832600176334381, | |
| "eval_precision": 0.8954019032094056, | |
| "eval_recall": 0.8949326095168356, | |
| "eval_runtime": 19.6658, | |
| "eval_samples_per_second": 87.054, | |
| "eval_steps_per_second": 5.441, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 1.8552567958831787, | |
| "learning_rate": 0.004491978609625669, | |
| "loss": 0.4441, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 2.53872013092041, | |
| "learning_rate": 0.00446524064171123, | |
| "loss": 0.4436, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 1.3826777935028076, | |
| "learning_rate": 0.004438502673796791, | |
| "loss": 0.4632, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 2.2216227054595947, | |
| "learning_rate": 0.004411764705882353, | |
| "loss": 0.4429, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 1.8521422147750854, | |
| "learning_rate": 0.004385026737967914, | |
| "loss": 0.4472, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 2.058058977127075, | |
| "learning_rate": 0.00436096256684492, | |
| "loss": 0.4757, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 1.1437183618545532, | |
| "learning_rate": 0.004334224598930481, | |
| "loss": 0.3436, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 1.761400580406189, | |
| "learning_rate": 0.0043074866310160425, | |
| "loss": 0.4958, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.4134129285812378, | |
| "learning_rate": 0.004280748663101605, | |
| "loss": 0.4519, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 1.7341545820236206, | |
| "learning_rate": 0.004254010695187166, | |
| "loss": 0.528, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 2.980020761489868, | |
| "learning_rate": 0.004227272727272727, | |
| "loss": 0.5021, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.6755030751228333, | |
| "learning_rate": 0.004200534759358289, | |
| "loss": 0.4601, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 1.8686202764511108, | |
| "learning_rate": 0.00417379679144385, | |
| "loss": 0.4433, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 1.371077299118042, | |
| "learning_rate": 0.004147058823529412, | |
| "loss": 0.4323, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 1.0771093368530273, | |
| "learning_rate": 0.004120320855614973, | |
| "loss": 0.4251, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 1.185023546218872, | |
| "learning_rate": 0.004093582887700535, | |
| "loss": 0.4881, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 0.9843281507492065, | |
| "learning_rate": 0.004066844919786096, | |
| "loss": 0.4483, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 1.6477869749069214, | |
| "learning_rate": 0.004040106951871658, | |
| "loss": 0.4956, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 1.7044633626937866, | |
| "learning_rate": 0.004013368983957219, | |
| "loss": 0.4176, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9310747663551402, | |
| "eval_f1": 0.9182291375322846, | |
| "eval_loss": 0.198581263422966, | |
| "eval_precision": 0.924344363515161, | |
| "eval_recall": 0.9209393532374213, | |
| "eval_runtime": 19.6732, | |
| "eval_samples_per_second": 87.022, | |
| "eval_steps_per_second": 5.439, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 1.662027359008789, | |
| "learning_rate": 0.003986631016042781, | |
| "loss": 0.4022, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 1.188351035118103, | |
| "learning_rate": 0.003959893048128342, | |
| "loss": 0.3758, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 2.2225048542022705, | |
| "learning_rate": 0.003933155080213904, | |
| "loss": 0.4491, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 1.683356761932373, | |
| "learning_rate": 0.0039064171122994654, | |
| "loss": 0.3647, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 1.7646687030792236, | |
| "learning_rate": 0.0038796791443850265, | |
| "loss": 0.4666, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 2.173644781112671, | |
| "learning_rate": 0.0038529411764705885, | |
| "loss": 0.4314, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 0.8064551949501038, | |
| "learning_rate": 0.00382620320855615, | |
| "loss": 0.3944, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 0.9698677062988281, | |
| "learning_rate": 0.003799465240641711, | |
| "loss": 0.4314, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 0.9321346879005432, | |
| "learning_rate": 0.0037727272727272726, | |
| "loss": 0.467, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 2.6592769622802734, | |
| "learning_rate": 0.003745989304812834, | |
| "loss": 0.4024, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 1.7124016284942627, | |
| "learning_rate": 0.003719251336898396, | |
| "loss": 0.3283, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 3.178034543991089, | |
| "learning_rate": 0.0036925133689839572, | |
| "loss": 0.4377, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 1.2681751251220703, | |
| "learning_rate": 0.0036657754010695188, | |
| "loss": 0.3866, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 1.1923668384552002, | |
| "learning_rate": 0.0036390374331550803, | |
| "loss": 0.3366, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 1.499803066253662, | |
| "learning_rate": 0.0036122994652406414, | |
| "loss": 0.4578, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 1.887222409248352, | |
| "learning_rate": 0.0035855614973262034, | |
| "loss": 0.4189, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 1.3592134714126587, | |
| "learning_rate": 0.003558823529411765, | |
| "loss": 0.4008, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 3.0257527828216553, | |
| "learning_rate": 0.0035320855614973264, | |
| "loss": 0.3774, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 1.093493103981018, | |
| "learning_rate": 0.0035053475935828875, | |
| "loss": 0.3454, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9503504672897196, | |
| "eval_f1": 0.9402807880914787, | |
| "eval_loss": 0.15674300491809845, | |
| "eval_precision": 0.9426615409260363, | |
| "eval_recall": 0.9397047025483766, | |
| "eval_runtime": 19.5644, | |
| "eval_samples_per_second": 87.506, | |
| "eval_steps_per_second": 5.469, | |
| "step": 561 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 1.7053543329238892, | |
| "learning_rate": 0.003478609625668449, | |
| "loss": 0.3776, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 1.5041882991790771, | |
| "learning_rate": 0.003451871657754011, | |
| "loss": 0.4058, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 1.3619967699050903, | |
| "learning_rate": 0.0034251336898395725, | |
| "loss": 0.3646, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "grad_norm": 1.1415998935699463, | |
| "learning_rate": 0.0033983957219251336, | |
| "loss": 0.4906, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "grad_norm": 1.6870795488357544, | |
| "learning_rate": 0.003371657754010695, | |
| "loss": 0.3828, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 1.0538561344146729, | |
| "learning_rate": 0.0033449197860962567, | |
| "loss": 0.3728, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "grad_norm": 2.340454339981079, | |
| "learning_rate": 0.0033181818181818186, | |
| "loss": 0.3809, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "grad_norm": 2.317230224609375, | |
| "learning_rate": 0.0032914438502673797, | |
| "loss": 0.3391, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 1.242281436920166, | |
| "learning_rate": 0.0032647058823529413, | |
| "loss": 0.4091, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "grad_norm": 1.23116934299469, | |
| "learning_rate": 0.003237967914438503, | |
| "loss": 0.3592, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "grad_norm": 1.117090106010437, | |
| "learning_rate": 0.003211229946524064, | |
| "loss": 0.3867, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 1.0917716026306152, | |
| "learning_rate": 0.0031844919786096254, | |
| "loss": 0.4386, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "grad_norm": 1.2080508470535278, | |
| "learning_rate": 0.0031577540106951874, | |
| "loss": 0.3466, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "grad_norm": 1.695580244064331, | |
| "learning_rate": 0.003131016042780749, | |
| "loss": 0.3147, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 1.1604491472244263, | |
| "learning_rate": 0.00310427807486631, | |
| "loss": 0.3585, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 1.8931636810302734, | |
| "learning_rate": 0.0030775401069518715, | |
| "loss": 0.3578, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 1.4620869159698486, | |
| "learning_rate": 0.003050802139037433, | |
| "loss": 0.3522, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 1.3944414854049683, | |
| "learning_rate": 0.003024064171122995, | |
| "loss": 0.3228, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9357476635514018, | |
| "eval_f1": 0.9283408808159395, | |
| "eval_loss": 0.1848856657743454, | |
| "eval_precision": 0.9231661406901872, | |
| "eval_recall": 0.9426484043891363, | |
| "eval_runtime": 19.683, | |
| "eval_samples_per_second": 86.979, | |
| "eval_steps_per_second": 5.436, | |
| "step": 748 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 1.7614619731903076, | |
| "learning_rate": 0.002997326203208556, | |
| "loss": 0.3463, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 2.866691827774048, | |
| "learning_rate": 0.0029705882352941177, | |
| "loss": 0.3431, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "grad_norm": 3.0871615409851074, | |
| "learning_rate": 0.002943850267379679, | |
| "loss": 0.4329, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 1.3399722576141357, | |
| "learning_rate": 0.0029171122994652403, | |
| "loss": 0.3992, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "grad_norm": 1.440559983253479, | |
| "learning_rate": 0.0028903743315508022, | |
| "loss": 0.3333, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "grad_norm": 1.4606270790100098, | |
| "learning_rate": 0.0028636363636363638, | |
| "loss": 0.3108, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "grad_norm": 2.4641544818878174, | |
| "learning_rate": 0.0028368983957219253, | |
| "loss": 0.3436, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "grad_norm": 1.9653208255767822, | |
| "learning_rate": 0.0028101604278074864, | |
| "loss": 0.2766, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "grad_norm": 1.0840091705322266, | |
| "learning_rate": 0.002783422459893048, | |
| "loss": 0.2568, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 1.0625332593917847, | |
| "learning_rate": 0.00275668449197861, | |
| "loss": 0.3366, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 0.9171143174171448, | |
| "learning_rate": 0.0027299465240641714, | |
| "loss": 0.339, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 1.6296868324279785, | |
| "learning_rate": 0.0027032085561497325, | |
| "loss": 0.359, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "grad_norm": 1.949312448501587, | |
| "learning_rate": 0.002676470588235294, | |
| "loss": 0.3529, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "grad_norm": 1.6241270303726196, | |
| "learning_rate": 0.0026497326203208556, | |
| "loss": 0.3364, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "grad_norm": 2.172145366668701, | |
| "learning_rate": 0.0026229946524064175, | |
| "loss": 0.3374, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "grad_norm": 3.377912998199463, | |
| "learning_rate": 0.0025962566844919786, | |
| "loss": 0.3555, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "grad_norm": 1.194082260131836, | |
| "learning_rate": 0.00256951871657754, | |
| "loss": 0.3354, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "grad_norm": 1.774932861328125, | |
| "learning_rate": 0.0025427807486631017, | |
| "loss": 0.3728, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "grad_norm": 0.9065486192703247, | |
| "learning_rate": 0.002516042780748663, | |
| "loss": 0.3382, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9398364485981309, | |
| "eval_f1": 0.9320964504504674, | |
| "eval_loss": 0.16266803443431854, | |
| "eval_precision": 0.9301560138584124, | |
| "eval_recall": 0.9396981551324834, | |
| "eval_runtime": 19.6531, | |
| "eval_samples_per_second": 87.111, | |
| "eval_steps_per_second": 5.444, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 0.8373203873634338, | |
| "learning_rate": 0.0024893048128342248, | |
| "loss": 0.3115, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "grad_norm": 1.6470876932144165, | |
| "learning_rate": 0.002462566844919786, | |
| "loss": 0.3746, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "grad_norm": 2.556999444961548, | |
| "learning_rate": 0.002435828877005348, | |
| "loss": 0.3411, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "grad_norm": 1.753217101097107, | |
| "learning_rate": 0.002409090909090909, | |
| "loss": 0.3095, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "grad_norm": 2.667759895324707, | |
| "learning_rate": 0.0023823529411764704, | |
| "loss": 0.3358, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "grad_norm": 1.6711212396621704, | |
| "learning_rate": 0.002355614973262032, | |
| "loss": 0.3263, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "grad_norm": 1.8793816566467285, | |
| "learning_rate": 0.0023288770053475935, | |
| "loss": 0.3245, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "grad_norm": 1.3059521913528442, | |
| "learning_rate": 0.002302139037433155, | |
| "loss": 0.2904, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "grad_norm": 1.765958309173584, | |
| "learning_rate": 0.0022754010695187166, | |
| "loss": 0.3424, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "grad_norm": 0.9322473406791687, | |
| "learning_rate": 0.002248663101604278, | |
| "loss": 0.3716, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "grad_norm": 2.082515239715576, | |
| "learning_rate": 0.0022219251336898396, | |
| "loss": 0.2967, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "grad_norm": 1.6903836727142334, | |
| "learning_rate": 0.002195187165775401, | |
| "loss": 0.3244, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "grad_norm": 1.1631466150283813, | |
| "learning_rate": 0.0021684491978609627, | |
| "loss": 0.3141, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "grad_norm": 2.086376428604126, | |
| "learning_rate": 0.002141711229946524, | |
| "loss": 0.3211, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "grad_norm": 1.709187626838684, | |
| "learning_rate": 0.0021149732620320857, | |
| "loss": 0.3039, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "grad_norm": 1.7365305423736572, | |
| "learning_rate": 0.0020882352941176473, | |
| "loss": 0.2937, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "grad_norm": 1.2648741006851196, | |
| "learning_rate": 0.0020614973262032084, | |
| "loss": 0.2951, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "grad_norm": 1.2121895551681519, | |
| "learning_rate": 0.00203475935828877, | |
| "loss": 0.242, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "grad_norm": 1.6397563219070435, | |
| "learning_rate": 0.0020080213903743314, | |
| "loss": 0.3363, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9509345794392523, | |
| "eval_f1": 0.9456282248184136, | |
| "eval_loss": 0.14138737320899963, | |
| "eval_precision": 0.9497944760971885, | |
| "eval_recall": 0.9441674024122191, | |
| "eval_runtime": 19.937, | |
| "eval_samples_per_second": 85.87, | |
| "eval_steps_per_second": 5.367, | |
| "step": 1122 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 1.3460767269134521, | |
| "learning_rate": 0.001981283422459893, | |
| "loss": 0.3134, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 1.2124683856964111, | |
| "learning_rate": 0.0019545454545454545, | |
| "loss": 0.3028, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "grad_norm": 0.8806934952735901, | |
| "learning_rate": 0.001927807486631016, | |
| "loss": 0.2589, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "grad_norm": 1.059187889099121, | |
| "learning_rate": 0.0019010695187165775, | |
| "loss": 0.2888, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "grad_norm": 2.5121827125549316, | |
| "learning_rate": 0.001874331550802139, | |
| "loss": 0.2741, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "grad_norm": 1.0052329301834106, | |
| "learning_rate": 0.0018475935828877006, | |
| "loss": 0.3519, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "grad_norm": 1.4301072359085083, | |
| "learning_rate": 0.0018208556149732621, | |
| "loss": 0.2937, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "grad_norm": 1.09031343460083, | |
| "learning_rate": 0.0017941176470588236, | |
| "loss": 0.2252, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "grad_norm": 1.9657083749771118, | |
| "learning_rate": 0.001767379679144385, | |
| "loss": 0.267, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "grad_norm": 3.7427196502685547, | |
| "learning_rate": 0.0017406417112299467, | |
| "loss": 0.2493, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "grad_norm": 1.7291096448898315, | |
| "learning_rate": 0.001713903743315508, | |
| "loss": 0.2558, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "grad_norm": 2.8834567070007324, | |
| "learning_rate": 0.0016871657754010698, | |
| "loss": 0.3167, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "grad_norm": 1.6702009439468384, | |
| "learning_rate": 0.001660427807486631, | |
| "loss": 0.274, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "grad_norm": 1.7623697519302368, | |
| "learning_rate": 0.0016336898395721924, | |
| "loss": 0.2481, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "grad_norm": 1.8855972290039062, | |
| "learning_rate": 0.0016069518716577541, | |
| "loss": 0.2424, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "grad_norm": 1.7909148931503296, | |
| "learning_rate": 0.0015802139037433154, | |
| "loss": 0.2361, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "grad_norm": 1.424047589302063, | |
| "learning_rate": 0.001553475935828877, | |
| "loss": 0.2834, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "grad_norm": 1.3470966815948486, | |
| "learning_rate": 0.0015267379679144385, | |
| "loss": 0.2981, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9544392523364486, | |
| "eval_f1": 0.9480272544883982, | |
| "eval_loss": 0.11172817647457123, | |
| "eval_precision": 0.9458066711610336, | |
| "eval_recall": 0.9541586489707353, | |
| "eval_runtime": 19.6986, | |
| "eval_samples_per_second": 86.91, | |
| "eval_steps_per_second": 5.432, | |
| "step": 1309 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "grad_norm": 1.9716545343399048, | |
| "learning_rate": 0.0015, | |
| "loss": 0.2591, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "grad_norm": 2.347787618637085, | |
| "learning_rate": 0.0014732620320855616, | |
| "loss": 0.2324, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "grad_norm": 1.5514649152755737, | |
| "learning_rate": 0.001446524064171123, | |
| "loss": 0.2163, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "grad_norm": 3.073544979095459, | |
| "learning_rate": 0.0014197860962566844, | |
| "loss": 0.2889, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "grad_norm": 1.5972115993499756, | |
| "learning_rate": 0.0013930481283422461, | |
| "loss": 0.2589, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "grad_norm": 1.8408401012420654, | |
| "learning_rate": 0.0013663101604278075, | |
| "loss": 0.2333, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "grad_norm": 1.3704335689544678, | |
| "learning_rate": 0.0013395721925133692, | |
| "loss": 0.2103, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "grad_norm": 3.6621859073638916, | |
| "learning_rate": 0.0013128342245989305, | |
| "loss": 0.2413, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "grad_norm": 1.345258355140686, | |
| "learning_rate": 0.0012860962566844918, | |
| "loss": 0.2444, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "grad_norm": 1.354202389717102, | |
| "learning_rate": 0.0012593582887700536, | |
| "loss": 0.2288, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "grad_norm": 0.983450174331665, | |
| "learning_rate": 0.0012326203208556149, | |
| "loss": 0.2995, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "grad_norm": 1.7251689434051514, | |
| "learning_rate": 0.0012058823529411764, | |
| "loss": 0.2898, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "grad_norm": 1.4366217851638794, | |
| "learning_rate": 0.001179144385026738, | |
| "loss": 0.2509, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "grad_norm": 1.6491020917892456, | |
| "learning_rate": 0.0011524064171122995, | |
| "loss": 0.2191, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "grad_norm": 1.4462454319000244, | |
| "learning_rate": 0.001125668449197861, | |
| "loss": 0.2307, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "grad_norm": 1.5503740310668945, | |
| "learning_rate": 0.0010989304812834225, | |
| "loss": 0.2167, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "grad_norm": 1.5065810680389404, | |
| "learning_rate": 0.001072192513368984, | |
| "loss": 0.3377, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "grad_norm": 1.3696374893188477, | |
| "learning_rate": 0.0010454545454545454, | |
| "loss": 0.24, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "grad_norm": 0.9576804041862488, | |
| "learning_rate": 0.001018716577540107, | |
| "loss": 0.2214, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9649532710280374, | |
| "eval_f1": 0.9609815836403053, | |
| "eval_loss": 0.11309263855218887, | |
| "eval_precision": 0.9642473014777337, | |
| "eval_recall": 0.9584474051621633, | |
| "eval_runtime": 19.6442, | |
| "eval_samples_per_second": 87.15, | |
| "eval_steps_per_second": 5.447, | |
| "step": 1496 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "grad_norm": 1.6305640935897827, | |
| "learning_rate": 0.0009919786096256684, | |
| "loss": 0.2645, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 1.0711798667907715, | |
| "learning_rate": 0.00096524064171123, | |
| "loss": 0.2063, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "grad_norm": 1.2606171369552612, | |
| "learning_rate": 0.0009385026737967915, | |
| "loss": 0.1904, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "grad_norm": 0.8554580807685852, | |
| "learning_rate": 0.0009117647058823529, | |
| "loss": 0.2078, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "grad_norm": 1.0638494491577148, | |
| "learning_rate": 0.0008850267379679144, | |
| "loss": 0.2129, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "grad_norm": 1.4322021007537842, | |
| "learning_rate": 0.000858288770053476, | |
| "loss": 0.2761, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "grad_norm": 1.2639697790145874, | |
| "learning_rate": 0.0008315508021390375, | |
| "loss": 0.1979, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "grad_norm": 1.108430027961731, | |
| "learning_rate": 0.0008048128342245989, | |
| "loss": 0.2051, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "grad_norm": 2.08953857421875, | |
| "learning_rate": 0.0007780748663101605, | |
| "loss": 0.2306, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 1.464694857597351, | |
| "learning_rate": 0.000751336898395722, | |
| "loss": 0.1992, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "grad_norm": 1.4773173332214355, | |
| "learning_rate": 0.0007245989304812835, | |
| "loss": 0.1764, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "grad_norm": 2.048029661178589, | |
| "learning_rate": 0.000697860962566845, | |
| "loss": 0.237, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "grad_norm": 1.0951212644577026, | |
| "learning_rate": 0.0006711229946524064, | |
| "loss": 0.1821, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "grad_norm": 1.084712028503418, | |
| "learning_rate": 0.0006443850267379679, | |
| "loss": 0.1947, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "grad_norm": 1.007285714149475, | |
| "learning_rate": 0.0006176470588235294, | |
| "loss": 0.2014, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "grad_norm": 1.0643844604492188, | |
| "learning_rate": 0.0005909090909090909, | |
| "loss": 0.2411, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "grad_norm": 2.0171964168548584, | |
| "learning_rate": 0.0005641711229946525, | |
| "loss": 0.2297, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "grad_norm": 0.8814995884895325, | |
| "learning_rate": 0.0005374331550802139, | |
| "loss": 0.2052, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "grad_norm": 1.338088035583496, | |
| "learning_rate": 0.0005106951871657754, | |
| "loss": 0.1928, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9649532710280374, | |
| "eval_f1": 0.9624133353031232, | |
| "eval_loss": 0.09664417803287506, | |
| "eval_precision": 0.9632215980141733, | |
| "eval_recall": 0.9628324486352646, | |
| "eval_runtime": 19.7505, | |
| "eval_samples_per_second": 86.681, | |
| "eval_steps_per_second": 5.418, | |
| "step": 1683 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "grad_norm": 1.1753814220428467, | |
| "learning_rate": 0.0004839572192513369, | |
| "loss": 0.1862, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 0.9707505702972412, | |
| "learning_rate": 0.0004572192513368984, | |
| "loss": 0.2182, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "grad_norm": 0.9967671632766724, | |
| "learning_rate": 0.0004304812834224599, | |
| "loss": 0.1923, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "grad_norm": 1.496031641960144, | |
| "learning_rate": 0.00040374331550802143, | |
| "loss": 0.2105, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 0.8774816393852234, | |
| "learning_rate": 0.00037700534759358285, | |
| "loss": 0.1969, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "grad_norm": 0.6063610315322876, | |
| "learning_rate": 0.0003502673796791444, | |
| "loss": 0.1577, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "grad_norm": 0.8216743469238281, | |
| "learning_rate": 0.0003235294117647059, | |
| "loss": 0.2064, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "grad_norm": 0.7338688373565674, | |
| "learning_rate": 0.0002967914438502674, | |
| "loss": 0.1793, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "grad_norm": 0.910650372505188, | |
| "learning_rate": 0.00027005347593582886, | |
| "loss": 0.194, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "grad_norm": 0.7778304219245911, | |
| "learning_rate": 0.00024331550802139036, | |
| "loss": 0.2203, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 1.0693227052688599, | |
| "learning_rate": 0.00021657754010695186, | |
| "loss": 0.1718, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "grad_norm": 1.4808011054992676, | |
| "learning_rate": 0.0001898395721925134, | |
| "loss": 0.1696, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "grad_norm": 0.8625634908676147, | |
| "learning_rate": 0.0001631016042780749, | |
| "loss": 0.1875, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "grad_norm": 1.1236218214035034, | |
| "learning_rate": 0.00013636363636363637, | |
| "loss": 0.1772, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "grad_norm": 1.027061939239502, | |
| "learning_rate": 0.00010962566844919787, | |
| "loss": 0.2274, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "grad_norm": 0.977976381778717, | |
| "learning_rate": 8.288770053475936e-05, | |
| "loss": 0.1672, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "grad_norm": 0.957969069480896, | |
| "learning_rate": 5.614973262032086e-05, | |
| "loss": 0.1966, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "grad_norm": 0.6182002425193787, | |
| "learning_rate": 2.9411764705882354e-05, | |
| "loss": 0.1546, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.8023217916488647, | |
| "learning_rate": 2.6737967914438504e-06, | |
| "loss": 0.1901, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9713785046728972, | |
| "eval_f1": 0.9692014832223894, | |
| "eval_loss": 0.07747028768062592, | |
| "eval_precision": 0.968992240300534, | |
| "eval_recall": 0.9698888041231651, | |
| "eval_runtime": 19.6225, | |
| "eval_samples_per_second": 87.247, | |
| "eval_steps_per_second": 5.453, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1870, | |
| "total_flos": 2.1188849626596557e+19, | |
| "train_loss": 0.34750947773775315, | |
| "train_runtime": 3122.2212, | |
| "train_samples_per_second": 38.303, | |
| "train_steps_per_second": 0.599 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1870, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2.1188849626596557e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |