| { | |
| "best_metric": 0.9789719626168224, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-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": 0.9260491132736206, | |
| "learning_rate": 0.004973262032085562, | |
| "loss": 1.5983, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 1.1307735443115234, | |
| "learning_rate": 0.004946524064171123, | |
| "loss": 0.9417, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.9537946581840515, | |
| "learning_rate": 0.004919786096256685, | |
| "loss": 0.7642, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.8597701191902161, | |
| "learning_rate": 0.004893048128342246, | |
| "loss": 0.6992, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 1.104675531387329, | |
| "learning_rate": 0.004866310160427808, | |
| "loss": 0.627, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.846555233001709, | |
| "learning_rate": 0.004839572192513369, | |
| "loss": 0.5047, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.423182487487793, | |
| "learning_rate": 0.004812834224598931, | |
| "loss": 0.5431, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.8424627780914307, | |
| "learning_rate": 0.004786096256684492, | |
| "loss": 0.5962, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.6608781814575195, | |
| "learning_rate": 0.004759358288770054, | |
| "loss": 0.4084, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 1.130247712135315, | |
| "learning_rate": 0.004732620320855615, | |
| "loss": 0.4932, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.6054658889770508, | |
| "learning_rate": 0.004705882352941177, | |
| "loss": 0.4684, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.8725093603134155, | |
| "learning_rate": 0.004679144385026738, | |
| "loss": 0.4429, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.6343618035316467, | |
| "learning_rate": 0.0046524064171123, | |
| "loss": 0.3952, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.9175045490264893, | |
| "learning_rate": 0.0046256684491978615, | |
| "loss": 0.4592, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.0295114517211914, | |
| "learning_rate": 0.004598930481283423, | |
| "loss": 0.4212, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.4232007563114166, | |
| "learning_rate": 0.004572192513368984, | |
| "loss": 0.4165, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 1.18360435962677, | |
| "learning_rate": 0.00454812834224599, | |
| "loss": 0.4245, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.7265322804450989, | |
| "learning_rate": 0.004521390374331551, | |
| "loss": 0.4059, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9310747663551402, | |
| "eval_f1": 0.9201346862223367, | |
| "eval_loss": 0.18775394558906555, | |
| "eval_precision": 0.913178148427007, | |
| "eval_recall": 0.9327948208695145, | |
| "eval_runtime": 9.5068, | |
| "eval_samples_per_second": 180.082, | |
| "eval_steps_per_second": 11.255, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 1.179917335510254, | |
| "learning_rate": 0.004494652406417113, | |
| "loss": 0.3646, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 1.1189391613006592, | |
| "learning_rate": 0.004467914438502674, | |
| "loss": 0.4339, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.8059839010238647, | |
| "learning_rate": 0.004441176470588235, | |
| "loss": 0.373, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 1.5934990644454956, | |
| "learning_rate": 0.004414438502673797, | |
| "loss": 0.4089, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 0.5738559365272522, | |
| "learning_rate": 0.004387700534759359, | |
| "loss": 0.4181, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.0053284168243408, | |
| "learning_rate": 0.00436096256684492, | |
| "loss": 0.354, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.6736829280853271, | |
| "learning_rate": 0.004334224598930481, | |
| "loss": 0.2862, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 0.7684084177017212, | |
| "learning_rate": 0.0043074866310160425, | |
| "loss": 0.3533, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.04612135887146, | |
| "learning_rate": 0.004280748663101605, | |
| "loss": 0.3654, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.7823394536972046, | |
| "learning_rate": 0.004254010695187166, | |
| "loss": 0.4385, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.9472429752349854, | |
| "learning_rate": 0.004227272727272727, | |
| "loss": 0.4417, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.889252245426178, | |
| "learning_rate": 0.004200534759358289, | |
| "loss": 0.3873, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 0.7252718806266785, | |
| "learning_rate": 0.00417379679144385, | |
| "loss": 0.3717, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 0.8687788844108582, | |
| "learning_rate": 0.004147058823529412, | |
| "loss": 0.3854, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.6197172999382019, | |
| "learning_rate": 0.004122994652406417, | |
| "loss": 0.3748, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 0.6506063342094421, | |
| "learning_rate": 0.004096256684491978, | |
| "loss": 0.2923, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 0.5267966389656067, | |
| "learning_rate": 0.00406951871657754, | |
| "loss": 0.4045, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 1.1251919269561768, | |
| "learning_rate": 0.004042780748663102, | |
| "loss": 0.3988, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 1.114890456199646, | |
| "learning_rate": 0.004016042780748663, | |
| "loss": 0.3796, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9082943925233645, | |
| "eval_f1": 0.886066241884805, | |
| "eval_loss": 0.27294662594795227, | |
| "eval_precision": 0.9131012141299326, | |
| "eval_recall": 0.887497540228883, | |
| "eval_runtime": 9.2331, | |
| "eval_samples_per_second": 185.419, | |
| "eval_steps_per_second": 11.589, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 0.7117612361907959, | |
| "learning_rate": 0.003989304812834224, | |
| "loss": 0.3724, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 0.9159232974052429, | |
| "learning_rate": 0.00396524064171123, | |
| "loss": 0.3155, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 0.6797966957092285, | |
| "learning_rate": 0.003938502673796792, | |
| "loss": 0.3531, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 0.912696361541748, | |
| "learning_rate": 0.003911764705882353, | |
| "loss": 0.2788, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 1.0336519479751587, | |
| "learning_rate": 0.0038850267379679144, | |
| "loss": 0.3692, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.8013398051261902, | |
| "learning_rate": 0.003858288770053476, | |
| "loss": 0.3561, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 0.6950948238372803, | |
| "learning_rate": 0.003831550802139038, | |
| "loss": 0.3295, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 0.7441625595092773, | |
| "learning_rate": 0.003804812834224599, | |
| "loss": 0.3285, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 4.745124816894531, | |
| "learning_rate": 0.0037780748663101605, | |
| "loss": 0.4162, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 1.3873414993286133, | |
| "learning_rate": 0.003751336898395722, | |
| "loss": 0.3424, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 0.7891167402267456, | |
| "learning_rate": 0.0037272727272727275, | |
| "loss": 0.3043, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 1.013873815536499, | |
| "learning_rate": 0.003700534759358289, | |
| "loss": 0.3754, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 0.9377150535583496, | |
| "learning_rate": 0.00367379679144385, | |
| "loss": 0.3675, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 2.7368648052215576, | |
| "learning_rate": 0.0036470588235294117, | |
| "loss": 0.2901, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 1.5487793684005737, | |
| "learning_rate": 0.0036203208556149736, | |
| "loss": 0.482, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 8.680522918701172, | |
| "learning_rate": 0.003593582887700535, | |
| "loss": 0.378, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 1.3777785301208496, | |
| "learning_rate": 0.0035668449197860962, | |
| "loss": 0.4919, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 2.1192550659179688, | |
| "learning_rate": 0.0035401069518716578, | |
| "loss": 0.3751, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 9.656478881835938, | |
| "learning_rate": 0.0035133689839572193, | |
| "loss": 0.424, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.866822429906542, | |
| "eval_f1": 0.8491520459723211, | |
| "eval_loss": 0.3701097071170807, | |
| "eval_precision": 0.8797339861417046, | |
| "eval_recall": 0.8520089249800192, | |
| "eval_runtime": 9.219, | |
| "eval_samples_per_second": 185.702, | |
| "eval_steps_per_second": 11.606, | |
| "step": 561 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 1.5421924591064453, | |
| "learning_rate": 0.0034866310160427804, | |
| "loss": 0.4643, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 0.9370782375335693, | |
| "learning_rate": 0.0034598930481283424, | |
| "loss": 0.4274, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 1.6456141471862793, | |
| "learning_rate": 0.003433155080213904, | |
| "loss": 0.3616, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "grad_norm": 1.2138258218765259, | |
| "learning_rate": 0.0034064171122994654, | |
| "loss": 0.4241, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "grad_norm": 0.8959400057792664, | |
| "learning_rate": 0.0033796791443850265, | |
| "loss": 0.3392, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 0.8747026324272156, | |
| "learning_rate": 0.003352941176470588, | |
| "loss": 0.3533, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "grad_norm": 1.7161656618118286, | |
| "learning_rate": 0.00332620320855615, | |
| "loss": 0.3407, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "grad_norm": 0.9229569435119629, | |
| "learning_rate": 0.0032994652406417115, | |
| "loss": 0.3098, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 0.9468969702720642, | |
| "learning_rate": 0.0032727272727272726, | |
| "loss": 0.3896, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "grad_norm": 1.4430208206176758, | |
| "learning_rate": 0.003245989304812834, | |
| "loss": 0.3395, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "grad_norm": 1.20052969455719, | |
| "learning_rate": 0.0032192513368983957, | |
| "loss": 0.3448, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 1.1726669073104858, | |
| "learning_rate": 0.0031925133689839577, | |
| "loss": 0.342, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "grad_norm": 0.7881722450256348, | |
| "learning_rate": 0.0031657754010695188, | |
| "loss": 0.301, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "grad_norm": 0.7960072159767151, | |
| "learning_rate": 0.0031390374331550803, | |
| "loss": 0.2633, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.964872419834137, | |
| "learning_rate": 0.003112299465240642, | |
| "loss": 0.2691, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 0.9894037246704102, | |
| "learning_rate": 0.003085561497326203, | |
| "loss": 0.2859, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 1.0027267932891846, | |
| "learning_rate": 0.003058823529411765, | |
| "loss": 0.3027, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 1.0325654745101929, | |
| "learning_rate": 0.0030320855614973264, | |
| "loss": 0.3141, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9380841121495327, | |
| "eval_f1": 0.9283105641226367, | |
| "eval_loss": 0.18485769629478455, | |
| "eval_precision": 0.9266830676466586, | |
| "eval_recall": 0.9336478146798447, | |
| "eval_runtime": 9.3787, | |
| "eval_samples_per_second": 182.542, | |
| "eval_steps_per_second": 11.409, | |
| "step": 748 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 1.263634443283081, | |
| "learning_rate": 0.003005347593582888, | |
| "loss": 0.3592, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 1.8158007860183716, | |
| "learning_rate": 0.002978609625668449, | |
| "loss": 0.364, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "grad_norm": 0.9459696412086487, | |
| "learning_rate": 0.0029518716577540106, | |
| "loss": 0.3587, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 0.7624779343605042, | |
| "learning_rate": 0.0029251336898395725, | |
| "loss": 0.304, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "grad_norm": 0.8625235557556152, | |
| "learning_rate": 0.002898395721925134, | |
| "loss": 0.2726, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "grad_norm": 0.962257444858551, | |
| "learning_rate": 0.002871657754010695, | |
| "loss": 0.2601, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "grad_norm": 0.6333624720573425, | |
| "learning_rate": 0.0028449197860962567, | |
| "loss": 0.3448, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "grad_norm": 1.3983910083770752, | |
| "learning_rate": 0.002818181818181818, | |
| "loss": 0.3202, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "grad_norm": 0.6626348495483398, | |
| "learning_rate": 0.00279144385026738, | |
| "loss": 0.2529, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 0.8221544027328491, | |
| "learning_rate": 0.0027647058823529413, | |
| "loss": 0.2523, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 0.7872591018676758, | |
| "learning_rate": 0.002737967914438503, | |
| "loss": 0.2832, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 1.50129234790802, | |
| "learning_rate": 0.0027112299465240643, | |
| "loss": 0.2912, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "grad_norm": 0.7471727728843689, | |
| "learning_rate": 0.0026844919786096254, | |
| "loss": 0.3097, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "grad_norm": 0.6078329086303711, | |
| "learning_rate": 0.002657754010695187, | |
| "loss": 0.2657, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "grad_norm": 0.8674110174179077, | |
| "learning_rate": 0.002631016042780749, | |
| "loss": 0.2633, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "grad_norm": 0.5421575307846069, | |
| "learning_rate": 0.0026042780748663104, | |
| "loss": 0.257, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "grad_norm": 1.314867377281189, | |
| "learning_rate": 0.0025775401069518715, | |
| "loss": 0.2688, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "grad_norm": 0.698221743106842, | |
| "learning_rate": 0.002550802139037433, | |
| "loss": 0.2506, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "grad_norm": 0.5437451004981995, | |
| "learning_rate": 0.0025240641711229946, | |
| "loss": 0.2553, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9643691588785047, | |
| "eval_f1": 0.9617344813251135, | |
| "eval_loss": 0.1074606254696846, | |
| "eval_precision": 0.9630090863077152, | |
| "eval_recall": 0.9611619604560873, | |
| "eval_runtime": 9.213, | |
| "eval_samples_per_second": 185.824, | |
| "eval_steps_per_second": 11.614, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 0.9639925956726074, | |
| "learning_rate": 0.002497326203208556, | |
| "loss": 0.2186, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "grad_norm": 1.0346194505691528, | |
| "learning_rate": 0.0024705882352941176, | |
| "loss": 0.3163, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "grad_norm": 0.9101438522338867, | |
| "learning_rate": 0.002443850267379679, | |
| "loss": 0.257, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "grad_norm": 0.9387779831886292, | |
| "learning_rate": 0.0024171122994652407, | |
| "loss": 0.2745, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "grad_norm": 1.3407084941864014, | |
| "learning_rate": 0.0023903743315508022, | |
| "loss": 0.2775, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "grad_norm": 0.7988283038139343, | |
| "learning_rate": 0.0023636363636363638, | |
| "loss": 0.2568, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "grad_norm": 0.8980028033256531, | |
| "learning_rate": 0.0023368983957219253, | |
| "loss": 0.296, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "grad_norm": 0.8847124576568604, | |
| "learning_rate": 0.002310160427807487, | |
| "loss": 0.2525, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "grad_norm": 1.3140696287155151, | |
| "learning_rate": 0.002283422459893048, | |
| "loss": 0.2967, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "grad_norm": 0.6774911284446716, | |
| "learning_rate": 0.00225668449197861, | |
| "loss": 0.2735, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "grad_norm": 0.9686025977134705, | |
| "learning_rate": 0.002229946524064171, | |
| "loss": 0.2415, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "grad_norm": 1.3379433155059814, | |
| "learning_rate": 0.0022032085561497325, | |
| "loss": 0.2656, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "grad_norm": 0.6908765435218811, | |
| "learning_rate": 0.002176470588235294, | |
| "loss": 0.2532, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "grad_norm": 0.8308853507041931, | |
| "learning_rate": 0.0021497326203208556, | |
| "loss": 0.2428, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "grad_norm": 1.2064207792282104, | |
| "learning_rate": 0.002122994652406417, | |
| "loss": 0.2989, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "grad_norm": 0.8376064896583557, | |
| "learning_rate": 0.0020962566844919786, | |
| "loss": 0.2061, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "grad_norm": 0.9363247156143188, | |
| "learning_rate": 0.00206951871657754, | |
| "loss": 0.2447, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "grad_norm": 7.874444007873535, | |
| "learning_rate": 0.0020427807486631017, | |
| "loss": 0.2254, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "grad_norm": 0.9535788297653198, | |
| "learning_rate": 0.002016042780748663, | |
| "loss": 0.2686, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9485981308411215, | |
| "eval_f1": 0.9488981890553403, | |
| "eval_loss": 0.16793404519557953, | |
| "eval_precision": 0.9560571498851578, | |
| "eval_recall": 0.9437216744429628, | |
| "eval_runtime": 9.2543, | |
| "eval_samples_per_second": 184.995, | |
| "eval_steps_per_second": 11.562, | |
| "step": 1122 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 0.9278040528297424, | |
| "learning_rate": 0.0019893048128342247, | |
| "loss": 0.256, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 1.0177885293960571, | |
| "learning_rate": 0.0019625668449197863, | |
| "loss": 0.2173, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "grad_norm": 0.5898217558860779, | |
| "learning_rate": 0.0019358288770053476, | |
| "loss": 0.2257, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "grad_norm": 5.235673904418945, | |
| "learning_rate": 0.0019090909090909091, | |
| "loss": 0.2388, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "grad_norm": 1.1271004676818848, | |
| "learning_rate": 0.0018823529411764706, | |
| "loss": 0.2544, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "grad_norm": 0.6136900186538696, | |
| "learning_rate": 0.001855614973262032, | |
| "loss": 0.2785, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "grad_norm": 0.9343350529670715, | |
| "learning_rate": 0.0018288770053475937, | |
| "loss": 0.2304, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "grad_norm": 0.7129714488983154, | |
| "learning_rate": 0.001802139037433155, | |
| "loss": 0.1709, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "grad_norm": 0.8645954132080078, | |
| "learning_rate": 0.0017754010695187168, | |
| "loss": 0.2099, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "grad_norm": 0.4692780375480652, | |
| "learning_rate": 0.001748663101604278, | |
| "loss": 0.1801, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "grad_norm": 1.1131465435028076, | |
| "learning_rate": 0.0017219251336898396, | |
| "loss": 0.2187, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "grad_norm": 1.0496641397476196, | |
| "learning_rate": 0.0016951871657754011, | |
| "loss": 0.2381, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "grad_norm": 0.7512268424034119, | |
| "learning_rate": 0.0016684491978609627, | |
| "loss": 0.2171, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "grad_norm": 0.9206662774085999, | |
| "learning_rate": 0.0016417112299465242, | |
| "loss": 0.1716, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "grad_norm": 1.044285535812378, | |
| "learning_rate": 0.0016149732620320857, | |
| "loss": 0.1996, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "grad_norm": 1.5523549318313599, | |
| "learning_rate": 0.001588235294117647, | |
| "loss": 0.198, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "grad_norm": 0.7654513120651245, | |
| "learning_rate": 0.0015614973262032088, | |
| "loss": 0.2341, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "grad_norm": 1.145663857460022, | |
| "learning_rate": 0.00153475935828877, | |
| "loss": 0.2556, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9661214953271028, | |
| "eval_f1": 0.9619479557860847, | |
| "eval_loss": 0.09340371936559677, | |
| "eval_precision": 0.9651383824240083, | |
| "eval_recall": 0.9598949442531882, | |
| "eval_runtime": 9.0216, | |
| "eval_samples_per_second": 189.767, | |
| "eval_steps_per_second": 11.86, | |
| "step": 1309 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "grad_norm": 0.8554219603538513, | |
| "learning_rate": 0.0015080213903743314, | |
| "loss": 0.237, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "grad_norm": 0.7055748701095581, | |
| "learning_rate": 0.0014812834224598931, | |
| "loss": 0.2317, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "grad_norm": 1.0891897678375244, | |
| "learning_rate": 0.0014545454545454545, | |
| "loss": 0.1723, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "grad_norm": 0.5554465651512146, | |
| "learning_rate": 0.0014278074866310162, | |
| "loss": 0.1986, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "grad_norm": 1.0232211351394653, | |
| "learning_rate": 0.0014010695187165775, | |
| "loss": 0.2222, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "grad_norm": 0.6204003095626831, | |
| "learning_rate": 0.001374331550802139, | |
| "loss": 0.1827, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "grad_norm": 0.7353977560997009, | |
| "learning_rate": 0.0013475935828877006, | |
| "loss": 0.1649, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "grad_norm": 0.734186053276062, | |
| "learning_rate": 0.001320855614973262, | |
| "loss": 0.194, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "grad_norm": 0.47959616780281067, | |
| "learning_rate": 0.0012941176470588236, | |
| "loss": 0.1763, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "grad_norm": 0.6939826607704163, | |
| "learning_rate": 0.0012673796791443852, | |
| "loss": 0.2286, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "grad_norm": 0.948558509349823, | |
| "learning_rate": 0.0012406417112299467, | |
| "loss": 0.2506, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "grad_norm": 0.8466843962669373, | |
| "learning_rate": 0.001213903743315508, | |
| "loss": 0.2175, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "grad_norm": 0.6146303415298462, | |
| "learning_rate": 0.0011871657754010695, | |
| "loss": 0.1641, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "grad_norm": 0.8321207761764526, | |
| "learning_rate": 0.001160427807486631, | |
| "loss": 0.1903, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "grad_norm": 0.7309682965278625, | |
| "learning_rate": 0.0011336898395721926, | |
| "loss": 0.1981, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "grad_norm": 0.5901007652282715, | |
| "learning_rate": 0.0011069518716577541, | |
| "loss": 0.2011, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "grad_norm": 0.9141890406608582, | |
| "learning_rate": 0.0010802139037433154, | |
| "loss": 0.2735, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "grad_norm": 0.813578724861145, | |
| "learning_rate": 0.001053475935828877, | |
| "loss": 0.2093, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "grad_norm": 0.4584049582481384, | |
| "learning_rate": 0.0010267379679144385, | |
| "loss": 0.1777, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.969626168224299, | |
| "eval_f1": 0.9686486797969157, | |
| "eval_loss": 0.08350867033004761, | |
| "eval_precision": 0.9696703038283683, | |
| "eval_recall": 0.9682591946397131, | |
| "eval_runtime": 9.2254, | |
| "eval_samples_per_second": 185.574, | |
| "eval_steps_per_second": 11.598, | |
| "step": 1496 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "grad_norm": 0.7080217599868774, | |
| "learning_rate": 0.001, | |
| "loss": 0.1999, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 0.9281997084617615, | |
| "learning_rate": 0.0009732620320855614, | |
| "loss": 0.1688, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "grad_norm": 0.8174493312835693, | |
| "learning_rate": 0.000946524064171123, | |
| "loss": 0.1731, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "grad_norm": 0.6349031925201416, | |
| "learning_rate": 0.0009197860962566845, | |
| "loss": 0.1672, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "grad_norm": 0.8174115419387817, | |
| "learning_rate": 0.000893048128342246, | |
| "loss": 0.1839, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "grad_norm": 0.6900407671928406, | |
| "learning_rate": 0.0008663101604278075, | |
| "loss": 0.2044, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "grad_norm": 0.2948859930038452, | |
| "learning_rate": 0.000839572192513369, | |
| "loss": 0.1328, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "grad_norm": 0.7020041942596436, | |
| "learning_rate": 0.0008128342245989305, | |
| "loss": 0.1759, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "grad_norm": 1.0418401956558228, | |
| "learning_rate": 0.000786096256684492, | |
| "loss": 0.1777, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 0.7473070025444031, | |
| "learning_rate": 0.0007593582887700536, | |
| "loss": 0.1631, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "grad_norm": 0.8006024360656738, | |
| "learning_rate": 0.000732620320855615, | |
| "loss": 0.1566, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "grad_norm": 1.0594407320022583, | |
| "learning_rate": 0.0007058823529411765, | |
| "loss": 0.184, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "grad_norm": 0.6014285087585449, | |
| "learning_rate": 0.000679144385026738, | |
| "loss": 0.1583, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "grad_norm": 0.6736869812011719, | |
| "learning_rate": 0.0006524064171122996, | |
| "loss": 0.1468, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "grad_norm": 0.6957813501358032, | |
| "learning_rate": 0.0006256684491978609, | |
| "loss": 0.1731, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "grad_norm": 0.5073075294494629, | |
| "learning_rate": 0.0005989304812834224, | |
| "loss": 0.176, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "grad_norm": 0.5485414862632751, | |
| "learning_rate": 0.000572192513368984, | |
| "loss": 0.1936, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "grad_norm": 0.8590062856674194, | |
| "learning_rate": 0.0005454545454545455, | |
| "loss": 0.1795, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "grad_norm": 0.49274083971977234, | |
| "learning_rate": 0.000518716577540107, | |
| "loss": 0.1607, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9772196261682243, | |
| "eval_f1": 0.9758896890562156, | |
| "eval_loss": 0.07392112910747528, | |
| "eval_precision": 0.9732910812266744, | |
| "eval_recall": 0.97920005624388, | |
| "eval_runtime": 9.2433, | |
| "eval_samples_per_second": 185.214, | |
| "eval_steps_per_second": 11.576, | |
| "step": 1683 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "grad_norm": 0.4997323751449585, | |
| "learning_rate": 0.0004919786096256684, | |
| "loss": 0.1352, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 0.5221167206764221, | |
| "learning_rate": 0.00046524064171122996, | |
| "loss": 0.1597, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "grad_norm": 0.6731162071228027, | |
| "learning_rate": 0.0004385026737967915, | |
| "loss": 0.1639, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "grad_norm": 0.5156794786453247, | |
| "learning_rate": 0.00041176470588235296, | |
| "loss": 0.1667, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 0.767203152179718, | |
| "learning_rate": 0.0003850267379679145, | |
| "loss": 0.1672, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "grad_norm": 0.5664710402488708, | |
| "learning_rate": 0.0003582887700534759, | |
| "loss": 0.1428, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "grad_norm": 0.37641459703445435, | |
| "learning_rate": 0.00033155080213903744, | |
| "loss": 0.1667, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "grad_norm": 0.5527117252349854, | |
| "learning_rate": 0.0003048128342245989, | |
| "loss": 0.1723, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "grad_norm": 0.8746387958526611, | |
| "learning_rate": 0.00027807486631016044, | |
| "loss": 0.1596, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "grad_norm": 0.5461722612380981, | |
| "learning_rate": 0.0002513368983957219, | |
| "loss": 0.17, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 0.5201784372329712, | |
| "learning_rate": 0.00022459893048128345, | |
| "loss": 0.1268, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "grad_norm": 0.44921737909317017, | |
| "learning_rate": 0.00019786096256684492, | |
| "loss": 0.1537, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "grad_norm": 0.6538177728652954, | |
| "learning_rate": 0.00017112299465240642, | |
| "loss": 0.1564, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "grad_norm": 0.39654332399368286, | |
| "learning_rate": 0.00014438502673796793, | |
| "loss": 0.1196, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "grad_norm": 0.5751528143882751, | |
| "learning_rate": 0.00011764705882352942, | |
| "loss": 0.1953, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "grad_norm": 0.7018762826919556, | |
| "learning_rate": 9.09090909090909e-05, | |
| "loss": 0.1414, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "grad_norm": 0.8955555558204651, | |
| "learning_rate": 6.41711229946524e-05, | |
| "loss": 0.1415, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "grad_norm": 0.29650095105171204, | |
| "learning_rate": 3.74331550802139e-05, | |
| "loss": 0.1361, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.6939311623573303, | |
| "learning_rate": 1.0695187165775402e-05, | |
| "loss": 0.1898, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9789719626168224, | |
| "eval_f1": 0.9786328578443323, | |
| "eval_loss": 0.06271301954984665, | |
| "eval_precision": 0.9764445771965571, | |
| "eval_recall": 0.9811556249771411, | |
| "eval_runtime": 8.9863, | |
| "eval_samples_per_second": 190.512, | |
| "eval_steps_per_second": 11.907, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1870, | |
| "total_flos": 9.332136680499118e+18, | |
| "train_loss": 0.29726991015959553, | |
| "train_runtime": 1395.7704, | |
| "train_samples_per_second": 85.68, | |
| "train_steps_per_second": 1.34 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1870, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 9.332136680499118e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |