diff --git "a/trainer_state.json" "b/trainer_state.json" --- "a/trainer_state.json" +++ "b/trainer_state.json" @@ -1,4333 +1,3397 @@ { - "best_metric": 0.9874, - "best_model_checkpoint": "SWIN-AI-Image-Detector/checkpoint-6329", - "epoch": 9.996445076430856, - "global_step": 7030, + "best_metric": 0.9833333333333333, + "best_model_checkpoint": "SWIN-AI-Image-Detector/checkpoint-3829", + "epoch": 10.0, + "global_step": 5470, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { - "epoch": 0.01, - "learning_rate": 7.112375533428166e-07, - "loss": 0.7628, + "epoch": 0.02, + "learning_rate": 9.140767824497258e-07, + "loss": 0.7202, "step": 10 }, { - "epoch": 0.03, - "learning_rate": 1.4224751066856332e-06, - "loss": 0.7346, + "epoch": 0.04, + "learning_rate": 1.8281535648994516e-06, + "loss": 0.7217, "step": 20 }, { - "epoch": 0.04, - "learning_rate": 2.1337126600284495e-06, - "loss": 0.7017, + "epoch": 0.05, + "learning_rate": 2.7422303473491773e-06, + "loss": 0.7037, "step": 30 }, { - "epoch": 0.06, - "learning_rate": 2.8449502133712664e-06, - "loss": 0.6932, + "epoch": 0.07, + "learning_rate": 3.6563071297989032e-06, + "loss": 0.6712, "step": 40 }, { - "epoch": 0.07, - "learning_rate": 3.556187766714083e-06, - "loss": 0.6677, + "epoch": 0.09, + "learning_rate": 4.570383912248629e-06, + "loss": 0.6445, "step": 50 }, { - "epoch": 0.09, - "learning_rate": 4.267425320056899e-06, - "loss": 0.648, + "epoch": 0.11, + "learning_rate": 5.484460694698355e-06, + "loss": 0.61, "step": 60 }, { - "epoch": 0.1, - "learning_rate": 4.9786628733997155e-06, - "loss": 0.6154, + "epoch": 0.13, + "learning_rate": 6.398537477148081e-06, + "loss": 0.5873, "step": 70 }, { - "epoch": 0.11, - "learning_rate": 5.689900426742533e-06, - "loss": 0.5817, + "epoch": 0.15, + "learning_rate": 7.3126142595978065e-06, + "loss": 0.5315, "step": 80 }, { - "epoch": 0.13, - "learning_rate": 6.401137980085349e-06, - "loss": 0.5484, + "epoch": 0.16, + "learning_rate": 8.226691042047533e-06, + "loss": 0.5032, "step": 90 }, { - "epoch": 0.14, - "learning_rate": 7.112375533428166e-06, - "loss": 0.5137, + "epoch": 0.18, + "learning_rate": 9.140767824497258e-06, + "loss": 0.4722, "step": 100 }, { - "epoch": 0.16, - "learning_rate": 7.823613086770983e-06, - "loss": 0.4486, + "epoch": 0.2, + "learning_rate": 1.0054844606946984e-05, + "loss": 0.4301, "step": 110 }, { - "epoch": 0.17, - "learning_rate": 8.534850640113798e-06, - "loss": 0.4474, + "epoch": 0.22, + "learning_rate": 1.096892138939671e-05, + "loss": 0.4045, "step": 120 }, { - "epoch": 0.18, - "learning_rate": 9.246088193456614e-06, - "loss": 0.4034, + "epoch": 0.24, + "learning_rate": 1.1882998171846435e-05, + "loss": 0.3834, "step": 130 }, { - "epoch": 0.2, - "learning_rate": 9.957325746799431e-06, - "loss": 0.4124, + "epoch": 0.26, + "learning_rate": 1.2797074954296162e-05, + "loss": 0.3638, "step": 140 }, { - "epoch": 0.21, - "learning_rate": 1.0668563300142247e-05, - "loss": 0.373, + "epoch": 0.27, + "learning_rate": 1.3711151736745886e-05, + "loss": 0.3989, "step": 150 }, { - "epoch": 0.23, - "learning_rate": 1.1379800853485066e-05, - "loss": 0.3743, + "epoch": 0.29, + "learning_rate": 1.4625228519195613e-05, + "loss": 0.3732, "step": 160 }, { - "epoch": 0.24, - "learning_rate": 1.2091038406827882e-05, - "loss": 0.3942, + "epoch": 0.31, + "learning_rate": 1.553930530164534e-05, + "loss": 0.3499, "step": 170 }, { - "epoch": 0.26, - "learning_rate": 1.2802275960170699e-05, - "loss": 0.3733, + "epoch": 0.33, + "learning_rate": 1.6453382084095066e-05, + "loss": 0.3414, "step": 180 }, { - "epoch": 0.27, - "learning_rate": 1.3513513513513515e-05, - "loss": 0.335, + "epoch": 0.35, + "learning_rate": 1.7367458866544793e-05, + "loss": 0.3469, "step": 190 }, { - "epoch": 0.28, - "learning_rate": 1.4224751066856332e-05, - "loss": 0.335, + "epoch": 0.37, + "learning_rate": 1.8281535648994517e-05, + "loss": 0.3556, "step": 200 }, { - "epoch": 0.3, - "learning_rate": 1.4935988620199148e-05, - "loss": 0.3517, + "epoch": 0.38, + "learning_rate": 1.9195612431444244e-05, + "loss": 0.3356, "step": 210 }, { - "epoch": 0.31, - "learning_rate": 1.5647226173541966e-05, - "loss": 0.3076, + "epoch": 0.4, + "learning_rate": 2.0109689213893968e-05, + "loss": 0.3244, "step": 220 }, { - "epoch": 0.33, - "learning_rate": 1.6358463726884778e-05, - "loss": 0.3263, + "epoch": 0.42, + "learning_rate": 2.1023765996343695e-05, + "loss": 0.3274, "step": 230 }, { - "epoch": 0.34, - "learning_rate": 1.7069701280227596e-05, - "loss": 0.3275, + "epoch": 0.44, + "learning_rate": 2.193784277879342e-05, + "loss": 0.301, "step": 240 }, { - "epoch": 0.36, - "learning_rate": 1.7780938833570414e-05, - "loss": 0.2999, + "epoch": 0.46, + "learning_rate": 2.2851919561243146e-05, + "loss": 0.3249, "step": 250 }, { - "epoch": 0.37, - "learning_rate": 1.849217638691323e-05, - "loss": 0.29, + "epoch": 0.48, + "learning_rate": 2.376599634369287e-05, + "loss": 0.2922, "step": 260 }, { - "epoch": 0.38, - "learning_rate": 1.9203413940256047e-05, - "loss": 0.3474, + "epoch": 0.49, + "learning_rate": 2.4680073126142597e-05, + "loss": 0.3002, "step": 270 }, { - "epoch": 0.4, - "learning_rate": 1.9914651493598862e-05, - "loss": 0.3244, + "epoch": 0.51, + "learning_rate": 2.5594149908592324e-05, + "loss": 0.2998, "step": 280 }, { - "epoch": 0.41, - "learning_rate": 2.062588904694168e-05, - "loss": 0.3155, + "epoch": 0.53, + "learning_rate": 2.6508226691042048e-05, + "loss": 0.3251, "step": 290 }, { - "epoch": 0.43, - "learning_rate": 2.1337126600284495e-05, - "loss": 0.2766, + "epoch": 0.55, + "learning_rate": 2.742230347349177e-05, + "loss": 0.3045, "step": 300 }, { - "epoch": 0.44, - "learning_rate": 2.2048364153627313e-05, - "loss": 0.2595, + "epoch": 0.57, + "learning_rate": 2.8336380255941502e-05, + "loss": 0.2822, "step": 310 }, { - "epoch": 0.46, - "learning_rate": 2.275960170697013e-05, - "loss": 0.3204, + "epoch": 0.59, + "learning_rate": 2.9250457038391226e-05, + "loss": 0.2582, "step": 320 }, { - "epoch": 0.47, - "learning_rate": 2.3470839260312946e-05, - "loss": 0.269, + "epoch": 0.6, + "learning_rate": 3.016453382084095e-05, + "loss": 0.3063, "step": 330 }, { - "epoch": 0.48, - "learning_rate": 2.4182076813655764e-05, - "loss": 0.2864, + "epoch": 0.62, + "learning_rate": 3.107861060329068e-05, + "loss": 0.2475, "step": 340 }, { - "epoch": 0.5, - "learning_rate": 2.489331436699858e-05, - "loss": 0.2387, + "epoch": 0.64, + "learning_rate": 3.1992687385740404e-05, + "loss": 0.3006, "step": 350 }, { - "epoch": 0.51, - "learning_rate": 2.5604551920341397e-05, - "loss": 0.2619, + "epoch": 0.66, + "learning_rate": 3.290676416819013e-05, + "loss": 0.3132, "step": 360 }, { - "epoch": 0.53, - "learning_rate": 2.6315789473684212e-05, - "loss": 0.3276, + "epoch": 0.68, + "learning_rate": 3.382084095063985e-05, + "loss": 0.2733, "step": 370 }, { - "epoch": 0.54, - "learning_rate": 2.702702702702703e-05, - "loss": 0.2843, + "epoch": 0.69, + "learning_rate": 3.4734917733089586e-05, + "loss": 0.2682, "step": 380 }, { - "epoch": 0.55, - "learning_rate": 2.7738264580369845e-05, - "loss": 0.2991, + "epoch": 0.71, + "learning_rate": 3.5648994515539306e-05, + "loss": 0.3016, "step": 390 }, { - "epoch": 0.57, - "learning_rate": 2.8449502133712663e-05, - "loss": 0.2723, + "epoch": 0.73, + "learning_rate": 3.656307129798903e-05, + "loss": 0.2722, "step": 400 }, { - "epoch": 0.58, - "learning_rate": 2.9160739687055478e-05, - "loss": 0.2446, + "epoch": 0.75, + "learning_rate": 3.7477148080438754e-05, + "loss": 0.2517, "step": 410 }, { - "epoch": 0.6, - "learning_rate": 2.9871977240398296e-05, - "loss": 0.25, + "epoch": 0.77, + "learning_rate": 3.839122486288849e-05, + "loss": 0.2823, "step": 420 }, { - "epoch": 0.61, - "learning_rate": 3.058321479374111e-05, - "loss": 0.2601, + "epoch": 0.79, + "learning_rate": 3.930530164533821e-05, + "loss": 0.2712, "step": 430 }, { - "epoch": 0.63, - "learning_rate": 3.129445234708393e-05, - "loss": 0.2581, + "epoch": 0.8, + "learning_rate": 4.0219378427787935e-05, + "loss": 0.297, "step": 440 }, { - "epoch": 0.64, - "learning_rate": 3.200568990042675e-05, - "loss": 0.2473, + "epoch": 0.82, + "learning_rate": 4.113345521023766e-05, + "loss": 0.2538, "step": 450 }, { - "epoch": 0.65, - "learning_rate": 3.2716927453769555e-05, - "loss": 0.278, + "epoch": 0.84, + "learning_rate": 4.204753199268739e-05, + "loss": 0.2803, "step": 460 }, { - "epoch": 0.67, - "learning_rate": 3.342816500711238e-05, - "loss": 0.289, + "epoch": 0.86, + "learning_rate": 4.296160877513711e-05, + "loss": 0.2442, "step": 470 }, { - "epoch": 0.68, - "learning_rate": 3.413940256045519e-05, - "loss": 0.2493, + "epoch": 0.88, + "learning_rate": 4.387568555758684e-05, + "loss": 0.2415, "step": 480 }, { - "epoch": 0.7, - "learning_rate": 3.485064011379801e-05, - "loss": 0.3039, + "epoch": 0.9, + "learning_rate": 4.4789762340036564e-05, + "loss": 0.2304, "step": 490 }, { - "epoch": 0.71, - "learning_rate": 3.556187766714083e-05, - "loss": 0.2833, + "epoch": 0.91, + "learning_rate": 4.570383912248629e-05, + "loss": 0.2743, "step": 500 }, { - "epoch": 0.73, - "learning_rate": 3.627311522048364e-05, - "loss": 0.2366, + "epoch": 0.93, + "learning_rate": 4.661791590493602e-05, + "loss": 0.2507, "step": 510 }, { - "epoch": 0.74, - "learning_rate": 3.698435277382646e-05, - "loss": 0.2751, + "epoch": 0.95, + "learning_rate": 4.753199268738574e-05, + "loss": 0.2488, "step": 520 }, { - "epoch": 0.75, - "learning_rate": 3.769559032716928e-05, - "loss": 0.2569, + "epoch": 0.97, + "learning_rate": 4.844606946983547e-05, + "loss": 0.2456, "step": 530 }, { - "epoch": 0.77, - "learning_rate": 3.8406827880512094e-05, - "loss": 0.2947, + "epoch": 0.99, + "learning_rate": 4.936014625228519e-05, + "loss": 0.2411, "step": 540 }, { - "epoch": 0.78, - "learning_rate": 3.911806543385491e-05, - "loss": 0.2916, + "epoch": 1.0, + "eval_accuracy": 0.9013333333333333, + "eval_loss": 0.24297115206718445, + "eval_runtime": 107.1641, + "eval_samples_per_second": 279.944, + "eval_steps_per_second": 8.753, + "step": 547 + }, + { + "epoch": 1.01, + "learning_rate": 4.996953077391834e-05, + "loss": 0.2475, "step": 550 }, { - "epoch": 0.8, - "learning_rate": 3.9829302987197724e-05, - "loss": 0.2456, + "epoch": 1.02, + "learning_rate": 4.986796668697948e-05, + "loss": 0.2483, "step": 560 }, { - "epoch": 0.81, - "learning_rate": 4.0540540540540545e-05, - "loss": 0.2592, + "epoch": 1.04, + "learning_rate": 4.976640260004063e-05, + "loss": 0.2289, "step": 570 }, { - "epoch": 0.82, - "learning_rate": 4.125177809388336e-05, - "loss": 0.2402, + "epoch": 1.06, + "learning_rate": 4.966483851310177e-05, + "loss": 0.2345, "step": 580 }, { - "epoch": 0.84, - "learning_rate": 4.1963015647226175e-05, - "loss": 0.2126, + "epoch": 1.08, + "learning_rate": 4.956327442616291e-05, + "loss": 0.2676, "step": 590 }, { - "epoch": 0.85, - "learning_rate": 4.267425320056899e-05, - "loss": 0.2507, + "epoch": 1.1, + "learning_rate": 4.946171033922405e-05, + "loss": 0.2422, "step": 600 }, { - "epoch": 0.87, - "learning_rate": 4.3385490753911804e-05, - "loss": 0.2062, + "epoch": 1.12, + "learning_rate": 4.936014625228519e-05, + "loss": 0.2348, "step": 610 }, { - "epoch": 0.88, - "learning_rate": 4.4096728307254626e-05, - "loss": 0.2415, + "epoch": 1.13, + "learning_rate": 4.9258582165346334e-05, + "loss": 0.2316, "step": 620 }, { - "epoch": 0.9, - "learning_rate": 4.480796586059744e-05, - "loss": 0.2459, + "epoch": 1.15, + "learning_rate": 4.915701807840748e-05, + "loss": 0.232, "step": 630 }, { - "epoch": 0.91, - "learning_rate": 4.551920341394026e-05, - "loss": 0.2641, + "epoch": 1.17, + "learning_rate": 4.905545399146862e-05, + "loss": 0.2271, "step": 640 }, { - "epoch": 0.92, - "learning_rate": 4.623044096728307e-05, - "loss": 0.2334, + "epoch": 1.19, + "learning_rate": 4.8953889904529756e-05, + "loss": 0.2086, "step": 650 }, { - "epoch": 0.94, - "learning_rate": 4.694167852062589e-05, - "loss": 0.255, + "epoch": 1.21, + "learning_rate": 4.8852325817590903e-05, + "loss": 0.2284, "step": 660 }, { - "epoch": 0.95, - "learning_rate": 4.765291607396871e-05, - "loss": 0.2355, + "epoch": 1.22, + "learning_rate": 4.8750761730652044e-05, + "loss": 0.2265, "step": 670 }, { - "epoch": 0.97, - "learning_rate": 4.836415362731153e-05, - "loss": 0.2416, + "epoch": 1.24, + "learning_rate": 4.8649197643713185e-05, + "loss": 0.227, "step": 680 }, { - "epoch": 0.98, - "learning_rate": 4.907539118065434e-05, - "loss": 0.2322, + "epoch": 1.26, + "learning_rate": 4.8547633556774326e-05, + "loss": 0.226, "step": 690 }, { - "epoch": 1.0, - "learning_rate": 4.978662873399716e-05, - "loss": 0.2432, + "epoch": 1.28, + "learning_rate": 4.844606946983547e-05, + "loss": 0.204, "step": 700 }, { - "epoch": 1.0, - "eval_accuracy": 0.9267, - "eval_loss": 0.1859399676322937, - "eval_runtime": 59.5879, - "eval_samples_per_second": 167.819, - "eval_steps_per_second": 5.253, - "step": 703 - }, - { - "epoch": 1.01, - "learning_rate": 4.9944681523628895e-05, - "loss": 0.2492, + "epoch": 1.3, + "learning_rate": 4.834450538289661e-05, + "loss": 0.2187, "step": 710 }, { - "epoch": 1.02, - "learning_rate": 4.9865655128813026e-05, - "loss": 0.2235, + "epoch": 1.32, + "learning_rate": 4.824294129595775e-05, + "loss": 0.2079, "step": 720 }, { - "epoch": 1.04, - "learning_rate": 4.978662873399716e-05, - "loss": 0.2463, + "epoch": 1.33, + "learning_rate": 4.8141377209018895e-05, + "loss": 0.2215, "step": 730 }, { - "epoch": 1.05, - "learning_rate": 4.970760233918128e-05, - "loss": 0.231, + "epoch": 1.35, + "learning_rate": 4.8039813122080036e-05, + "loss": 0.206, "step": 740 }, { - "epoch": 1.07, - "learning_rate": 4.962857594436542e-05, - "loss": 0.2507, + "epoch": 1.37, + "learning_rate": 4.7938249035141176e-05, + "loss": 0.2405, "step": 750 }, { - "epoch": 1.08, - "learning_rate": 4.954954954954955e-05, - "loss": 0.228, + "epoch": 1.39, + "learning_rate": 4.7836684948202324e-05, + "loss": 0.2155, "step": 760 }, { - "epoch": 1.09, - "learning_rate": 4.9470523154733685e-05, - "loss": 0.2438, + "epoch": 1.41, + "learning_rate": 4.773512086126346e-05, + "loss": 0.1981, "step": 770 }, { - "epoch": 1.11, - "learning_rate": 4.9391496759917816e-05, - "loss": 0.2249, + "epoch": 1.43, + "learning_rate": 4.76335567743246e-05, + "loss": 0.2124, "step": 780 }, { - "epoch": 1.12, - "learning_rate": 4.931247036510195e-05, - "loss": 0.2223, + "epoch": 1.44, + "learning_rate": 4.753199268738574e-05, + "loss": 0.2483, "step": 790 }, { - "epoch": 1.14, - "learning_rate": 4.923344397028608e-05, - "loss": 0.2163, + "epoch": 1.46, + "learning_rate": 4.7430428600446886e-05, + "loss": 0.2463, "step": 800 }, { - "epoch": 1.15, - "learning_rate": 4.9154417575470205e-05, - "loss": 0.2265, + "epoch": 1.48, + "learning_rate": 4.732886451350803e-05, + "loss": 0.2262, "step": 810 }, { - "epoch": 1.17, - "learning_rate": 4.907539118065434e-05, - "loss": 0.2032, + "epoch": 1.5, + "learning_rate": 4.722730042656917e-05, + "loss": 0.1905, "step": 820 }, { - "epoch": 1.18, - "learning_rate": 4.8996364785838475e-05, - "loss": 0.1977, + "epoch": 1.52, + "learning_rate": 4.712573633963031e-05, + "loss": 0.213, "step": 830 }, { - "epoch": 1.19, - "learning_rate": 4.8917338391022607e-05, - "loss": 0.2326, + "epoch": 1.54, + "learning_rate": 4.702417225269145e-05, + "loss": 0.1902, "step": 840 }, { - "epoch": 1.21, - "learning_rate": 4.883831199620674e-05, - "loss": 0.1894, + "epoch": 1.55, + "learning_rate": 4.692260816575259e-05, + "loss": 0.18, "step": 850 }, { - "epoch": 1.22, - "learning_rate": 4.875928560139087e-05, - "loss": 0.2655, + "epoch": 1.57, + "learning_rate": 4.682104407881374e-05, + "loss": 0.2204, "step": 860 }, { - "epoch": 1.24, - "learning_rate": 4.8680259206574995e-05, - "loss": 0.2196, + "epoch": 1.59, + "learning_rate": 4.671947999187488e-05, + "loss": 0.2002, "step": 870 }, { - "epoch": 1.25, - "learning_rate": 4.8601232811759127e-05, - "loss": 0.1939, + "epoch": 1.61, + "learning_rate": 4.661791590493602e-05, + "loss": 0.2229, "step": 880 }, { - "epoch": 1.27, - "learning_rate": 4.852220641694326e-05, - "loss": 0.2548, + "epoch": 1.63, + "learning_rate": 4.651635181799715e-05, + "loss": 0.1787, "step": 890 }, { - "epoch": 1.28, - "learning_rate": 4.84431800221274e-05, - "loss": 0.1991, + "epoch": 1.65, + "learning_rate": 4.64147877310583e-05, + "loss": 0.2312, "step": 900 }, { - "epoch": 1.29, - "learning_rate": 4.836415362731153e-05, - "loss": 0.1932, + "epoch": 1.66, + "learning_rate": 4.631322364411944e-05, + "loss": 0.2002, "step": 910 }, { - "epoch": 1.31, - "learning_rate": 4.828512723249566e-05, - "loss": 0.2012, + "epoch": 1.68, + "learning_rate": 4.621165955718058e-05, + "loss": 0.2167, "step": 920 }, { - "epoch": 1.32, - "learning_rate": 4.8206100837679785e-05, - "loss": 0.2109, + "epoch": 1.7, + "learning_rate": 4.611009547024173e-05, + "loss": 0.1949, "step": 930 }, { - "epoch": 1.34, - "learning_rate": 4.812707444286392e-05, - "loss": 0.199, + "epoch": 1.72, + "learning_rate": 4.600853138330287e-05, + "loss": 0.2002, "step": 940 }, { - "epoch": 1.35, - "learning_rate": 4.804804804804805e-05, - "loss": 0.2195, + "epoch": 1.74, + "learning_rate": 4.5906967296364e-05, + "loss": 0.1973, "step": 950 }, { - "epoch": 1.37, - "learning_rate": 4.796902165323218e-05, - "loss": 0.2223, + "epoch": 1.76, + "learning_rate": 4.580540320942515e-05, + "loss": 0.2049, "step": 960 }, { - "epoch": 1.38, - "learning_rate": 4.788999525841631e-05, - "loss": 0.1989, + "epoch": 1.77, + "learning_rate": 4.570383912248629e-05, + "loss": 0.1919, "step": 970 }, { - "epoch": 1.39, - "learning_rate": 4.781096886360045e-05, - "loss": 0.1844, + "epoch": 1.79, + "learning_rate": 4.560227503554743e-05, + "loss": 0.2147, "step": 980 }, { - "epoch": 1.41, - "learning_rate": 4.7731942468784575e-05, - "loss": 0.1992, + "epoch": 1.81, + "learning_rate": 4.550071094860857e-05, + "loss": 0.1964, "step": 990 }, { - "epoch": 1.42, - "learning_rate": 4.765291607396871e-05, - "loss": 0.1832, + "epoch": 1.83, + "learning_rate": 4.539914686166972e-05, + "loss": 0.1764, "step": 1000 }, { - "epoch": 1.44, - "learning_rate": 4.757388967915284e-05, - "loss": 0.207, + "epoch": 1.85, + "learning_rate": 4.5297582774730854e-05, + "loss": 0.1741, "step": 1010 }, { - "epoch": 1.45, - "learning_rate": 4.749486328433697e-05, - "loss": 0.2247, + "epoch": 1.86, + "learning_rate": 4.5196018687791995e-05, + "loss": 0.1963, "step": 1020 }, { - "epoch": 1.46, - "learning_rate": 4.74158368895211e-05, - "loss": 0.2278, + "epoch": 1.88, + "learning_rate": 4.509445460085314e-05, + "loss": 0.2276, "step": 1030 }, { - "epoch": 1.48, - "learning_rate": 4.7336810494705234e-05, - "loss": 0.2058, + "epoch": 1.9, + "learning_rate": 4.499289051391428e-05, + "loss": 0.1977, "step": 1040 }, { - "epoch": 1.49, - "learning_rate": 4.7257784099889365e-05, - "loss": 0.1964, + "epoch": 1.92, + "learning_rate": 4.4891326426975424e-05, + "loss": 0.1676, "step": 1050 }, { - "epoch": 1.51, - "learning_rate": 4.71787577050735e-05, - "loss": 0.186, + "epoch": 1.94, + "learning_rate": 4.4789762340036564e-05, + "loss": 0.1924, "step": 1060 }, { - "epoch": 1.52, - "learning_rate": 4.709973131025763e-05, - "loss": 0.2042, + "epoch": 1.96, + "learning_rate": 4.4688198253097705e-05, + "loss": 0.2131, "step": 1070 }, { - "epoch": 1.54, - "learning_rate": 4.702070491544176e-05, - "loss": 0.1665, + "epoch": 1.97, + "learning_rate": 4.4586634166158846e-05, + "loss": 0.2047, "step": 1080 }, { - "epoch": 1.55, - "learning_rate": 4.694167852062589e-05, - "loss": 0.2434, + "epoch": 1.99, + "learning_rate": 4.448507007921999e-05, + "loss": 0.2011, "step": 1090 }, { - "epoch": 1.56, - "learning_rate": 4.6862652125810024e-05, - "loss": 0.2005, + "epoch": 2.0, + "eval_accuracy": 0.9592666666666667, + "eval_loss": 0.10533071309328079, + "eval_runtime": 107.4335, + "eval_samples_per_second": 279.243, + "eval_steps_per_second": 8.731, + "step": 1094 + }, + { + "epoch": 2.01, + "learning_rate": 4.4383505992281134e-05, + "loss": 0.1827, "step": 1100 }, { - "epoch": 1.58, - "learning_rate": 4.678362573099415e-05, - "loss": 0.2025, + "epoch": 2.03, + "learning_rate": 4.4281941905342274e-05, + "loss": 0.1878, "step": 1110 }, { - "epoch": 1.59, - "learning_rate": 4.670459933617829e-05, - "loss": 0.189, + "epoch": 2.05, + "learning_rate": 4.4180377818403415e-05, + "loss": 0.1796, "step": 1120 }, { - "epoch": 1.61, - "learning_rate": 4.662557294136242e-05, - "loss": 0.1944, + "epoch": 2.07, + "learning_rate": 4.4078813731464556e-05, + "loss": 0.1562, "step": 1130 }, { - "epoch": 1.62, - "learning_rate": 4.654654654654655e-05, - "loss": 0.1941, + "epoch": 2.08, + "learning_rate": 4.3977249644525696e-05, + "loss": 0.1898, "step": 1140 }, { - "epoch": 1.64, - "learning_rate": 4.646752015173068e-05, - "loss": 0.1969, + "epoch": 2.1, + "learning_rate": 4.387568555758684e-05, + "loss": 0.1798, "step": 1150 }, { - "epoch": 1.65, - "learning_rate": 4.6388493756914814e-05, - "loss": 0.1794, + "epoch": 2.12, + "learning_rate": 4.3774121470647984e-05, + "loss": 0.1691, "step": 1160 }, { - "epoch": 1.66, - "learning_rate": 4.630946736209894e-05, - "loss": 0.1907, + "epoch": 2.14, + "learning_rate": 4.3672557383709125e-05, + "loss": 0.2052, "step": 1170 }, { - "epoch": 1.68, - "learning_rate": 4.623044096728307e-05, - "loss": 0.1809, + "epoch": 2.16, + "learning_rate": 4.3570993296770266e-05, + "loss": 0.1673, "step": 1180 }, { - "epoch": 1.69, - "learning_rate": 4.61514145724672e-05, - "loss": 0.1937, + "epoch": 2.18, + "learning_rate": 4.3469429209831407e-05, + "loss": 0.1788, "step": 1190 }, { - "epoch": 1.71, - "learning_rate": 4.607238817765134e-05, - "loss": 0.1916, + "epoch": 2.19, + "learning_rate": 4.336786512289255e-05, + "loss": 0.1986, "step": 1200 }, { - "epoch": 1.72, - "learning_rate": 4.599336178283547e-05, - "loss": 0.2069, + "epoch": 2.21, + "learning_rate": 4.326630103595369e-05, + "loss": 0.1988, "step": 1210 }, { - "epoch": 1.73, - "learning_rate": 4.5914335388019604e-05, - "loss": 0.1825, + "epoch": 2.23, + "learning_rate": 4.316473694901483e-05, + "loss": 0.1873, "step": 1220 }, { - "epoch": 1.75, - "learning_rate": 4.5835308993203736e-05, - "loss": 0.1821, + "epoch": 2.25, + "learning_rate": 4.3063172862075976e-05, + "loss": 0.1555, "step": 1230 }, { - "epoch": 1.76, - "learning_rate": 4.575628259838786e-05, - "loss": 0.1858, + "epoch": 2.27, + "learning_rate": 4.296160877513711e-05, + "loss": 0.1616, "step": 1240 }, { - "epoch": 1.78, - "learning_rate": 4.567725620357199e-05, - "loss": 0.174, + "epoch": 2.29, + "learning_rate": 4.286004468819825e-05, + "loss": 0.1578, "step": 1250 }, { - "epoch": 1.79, - "learning_rate": 4.5598229808756124e-05, - "loss": 0.1953, + "epoch": 2.3, + "learning_rate": 4.27584806012594e-05, + "loss": 0.1689, "step": 1260 }, { - "epoch": 1.81, - "learning_rate": 4.551920341394026e-05, - "loss": 0.1653, + "epoch": 2.32, + "learning_rate": 4.265691651432054e-05, + "loss": 0.1566, "step": 1270 }, { - "epoch": 1.82, - "learning_rate": 4.5440177019124394e-05, - "loss": 0.2142, + "epoch": 2.34, + "learning_rate": 4.255535242738168e-05, + "loss": 0.1789, "step": 1280 }, { - "epoch": 1.83, - "learning_rate": 4.5361150624308526e-05, - "loss": 0.194, + "epoch": 2.36, + "learning_rate": 4.245378834044283e-05, + "loss": 0.1798, "step": 1290 }, { - "epoch": 1.85, - "learning_rate": 4.528212422949265e-05, - "loss": 0.2183, + "epoch": 2.38, + "learning_rate": 4.235222425350396e-05, + "loss": 0.1831, "step": 1300 }, { - "epoch": 1.86, - "learning_rate": 4.520309783467678e-05, - "loss": 0.1974, + "epoch": 2.39, + "learning_rate": 4.22506601665651e-05, + "loss": 0.1817, "step": 1310 }, { - "epoch": 1.88, - "learning_rate": 4.5124071439860914e-05, - "loss": 0.171, + "epoch": 2.41, + "learning_rate": 4.214909607962624e-05, + "loss": 0.1636, "step": 1320 }, { - "epoch": 1.89, - "learning_rate": 4.5045045045045046e-05, - "loss": 0.158, + "epoch": 2.43, + "learning_rate": 4.204753199268739e-05, + "loss": 0.2407, "step": 1330 }, { - "epoch": 1.91, - "learning_rate": 4.496601865022918e-05, - "loss": 0.1823, + "epoch": 2.45, + "learning_rate": 4.194596790574853e-05, + "loss": 0.1733, "step": 1340 }, { - "epoch": 1.92, - "learning_rate": 4.4886992255413316e-05, - "loss": 0.1649, + "epoch": 2.47, + "learning_rate": 4.184440381880967e-05, + "loss": 0.1833, "step": 1350 }, { - "epoch": 1.93, - "learning_rate": 4.480796586059744e-05, - "loss": 0.1724, + "epoch": 2.49, + "learning_rate": 4.174283973187081e-05, + "loss": 0.1783, "step": 1360 }, { - "epoch": 1.95, - "learning_rate": 4.472893946578157e-05, - "loss": 0.1832, + "epoch": 2.5, + "learning_rate": 4.164127564493195e-05, + "loss": 0.1725, "step": 1370 }, { - "epoch": 1.96, - "learning_rate": 4.4649913070965704e-05, - "loss": 0.1657, + "epoch": 2.52, + "learning_rate": 4.153971155799309e-05, + "loss": 0.1401, "step": 1380 }, { - "epoch": 1.98, - "learning_rate": 4.4570886676149836e-05, - "loss": 0.1956, + "epoch": 2.54, + "learning_rate": 4.143814747105424e-05, + "loss": 0.1597, "step": 1390 }, { - "epoch": 1.99, - "learning_rate": 4.449186028133397e-05, - "loss": 0.1665, + "epoch": 2.56, + "learning_rate": 4.133658338411538e-05, + "loss": 0.1399, "step": 1400 }, { - "epoch": 2.0, - "eval_accuracy": 0.9514, - "eval_loss": 0.1229332983493805, - "eval_runtime": 59.6111, - "eval_samples_per_second": 167.754, - "eval_steps_per_second": 5.251, - "step": 1406 - }, - { - "epoch": 2.0, - "learning_rate": 4.44128338865181e-05, - "loss": 0.1707, + "epoch": 2.58, + "learning_rate": 4.123501929717652e-05, + "loss": 0.1735, "step": 1410 }, { - "epoch": 2.02, - "learning_rate": 4.433380749170223e-05, - "loss": 0.1837, + "epoch": 2.6, + "learning_rate": 4.113345521023766e-05, + "loss": 0.173, "step": 1420 }, { - "epoch": 2.03, - "learning_rate": 4.425478109688636e-05, - "loss": 0.1625, + "epoch": 2.61, + "learning_rate": 4.10318911232988e-05, + "loss": 0.1778, "step": 1430 }, { - "epoch": 2.05, - "learning_rate": 4.4175754702070494e-05, - "loss": 0.1683, + "epoch": 2.63, + "learning_rate": 4.0930327036359944e-05, + "loss": 0.1637, "step": 1440 }, { - "epoch": 2.06, - "learning_rate": 4.4096728307254626e-05, - "loss": 0.1493, + "epoch": 2.65, + "learning_rate": 4.0828762949421084e-05, + "loss": 0.1724, "step": 1450 }, { - "epoch": 2.08, - "learning_rate": 4.401770191243876e-05, - "loss": 0.1613, + "epoch": 2.67, + "learning_rate": 4.072719886248223e-05, + "loss": 0.161, "step": 1460 }, { - "epoch": 2.09, - "learning_rate": 4.393867551762289e-05, - "loss": 0.1741, + "epoch": 2.69, + "learning_rate": 4.062563477554337e-05, + "loss": 0.1609, "step": 1470 }, { - "epoch": 2.1, - "learning_rate": 4.3859649122807014e-05, - "loss": 0.1384, + "epoch": 2.71, + "learning_rate": 4.0524070688604506e-05, + "loss": 0.1719, "step": 1480 }, { - "epoch": 2.12, - "learning_rate": 4.378062272799115e-05, - "loss": 0.2032, + "epoch": 2.72, + "learning_rate": 4.0422506601665654e-05, + "loss": 0.1711, "step": 1490 }, { - "epoch": 2.13, - "learning_rate": 4.3701596333175284e-05, - "loss": 0.1721, + "epoch": 2.74, + "learning_rate": 4.0320942514726794e-05, + "loss": 0.1687, "step": 1500 }, { - "epoch": 2.15, - "learning_rate": 4.3622569938359416e-05, - "loss": 0.1664, + "epoch": 2.76, + "learning_rate": 4.0219378427787935e-05, + "loss": 0.1996, "step": 1510 }, { - "epoch": 2.16, - "learning_rate": 4.354354354354355e-05, - "loss": 0.1784, + "epoch": 2.78, + "learning_rate": 4.011781434084908e-05, + "loss": 0.144, "step": 1520 }, { - "epoch": 2.18, - "learning_rate": 4.346451714872768e-05, - "loss": 0.184, + "epoch": 2.8, + "learning_rate": 4.001625025391022e-05, + "loss": 0.1861, "step": 1530 }, { - "epoch": 2.19, - "learning_rate": 4.3385490753911804e-05, - "loss": 0.1558, + "epoch": 2.82, + "learning_rate": 3.991468616697136e-05, + "loss": 0.1492, "step": 1540 }, { - "epoch": 2.2, - "learning_rate": 4.3306464359095936e-05, - "loss": 0.1746, + "epoch": 2.83, + "learning_rate": 3.98131220800325e-05, + "loss": 0.1399, "step": 1550 }, { - "epoch": 2.22, - "learning_rate": 4.322743796428007e-05, - "loss": 0.1595, + "epoch": 2.85, + "learning_rate": 3.9711557993093645e-05, + "loss": 0.1488, "step": 1560 }, { - "epoch": 2.23, - "learning_rate": 4.3148411569464206e-05, - "loss": 0.1764, + "epoch": 2.87, + "learning_rate": 3.9609993906154786e-05, + "loss": 0.2207, "step": 1570 }, { - "epoch": 2.25, - "learning_rate": 4.306938517464834e-05, - "loss": 0.1575, + "epoch": 2.89, + "learning_rate": 3.9508429819215927e-05, + "loss": 0.1679, "step": 1580 }, { - "epoch": 2.26, - "learning_rate": 4.299035877983247e-05, - "loss": 0.173, + "epoch": 2.91, + "learning_rate": 3.9406865732277074e-05, + "loss": 0.159, "step": 1590 }, { - "epoch": 2.28, - "learning_rate": 4.2911332385016595e-05, - "loss": 0.1879, + "epoch": 2.93, + "learning_rate": 3.930530164533821e-05, + "loss": 0.1415, "step": 1600 }, { - "epoch": 2.29, - "learning_rate": 4.2832305990200726e-05, - "loss": 0.1879, + "epoch": 2.94, + "learning_rate": 3.920373755839935e-05, + "loss": 0.1436, "step": 1610 }, { - "epoch": 2.3, - "learning_rate": 4.275327959538486e-05, - "loss": 0.1313, + "epoch": 2.96, + "learning_rate": 3.9102173471460496e-05, + "loss": 0.1503, "step": 1620 }, { - "epoch": 2.32, - "learning_rate": 4.267425320056899e-05, - "loss": 0.1741, + "epoch": 2.98, + "learning_rate": 3.900060938452164e-05, + "loss": 0.2036, "step": 1630 }, { - "epoch": 2.33, - "learning_rate": 4.259522680575312e-05, - "loss": 0.1393, + "epoch": 3.0, + "learning_rate": 3.889904529758278e-05, + "loss": 0.1722, "step": 1640 }, { - "epoch": 2.35, - "learning_rate": 4.251620041093726e-05, - "loss": 0.164, - "step": 1650 + "epoch": 3.0, + "eval_accuracy": 0.9671333333333333, + "eval_loss": 0.08246280997991562, + "eval_runtime": 107.3836, + "eval_samples_per_second": 279.372, + "eval_steps_per_second": 8.735, + "step": 1641 }, { - "epoch": 2.36, - "learning_rate": 4.243717401612139e-05, - "loss": 0.1709, + "epoch": 3.02, + "learning_rate": 3.879748121064392e-05, + "loss": 0.1468, + "step": 1650 + }, + { + "epoch": 3.03, + "learning_rate": 3.869591712370506e-05, + "loss": 0.1678, "step": 1660 }, { - "epoch": 2.37, - "learning_rate": 4.2358147621305516e-05, - "loss": 0.1686, + "epoch": 3.05, + "learning_rate": 3.85943530367662e-05, + "loss": 0.1453, "step": 1670 }, { - "epoch": 2.39, - "learning_rate": 4.227912122648965e-05, - "loss": 0.1595, + "epoch": 3.07, + "learning_rate": 3.849278894982734e-05, + "loss": 0.1664, "step": 1680 }, { - "epoch": 2.4, - "learning_rate": 4.220009483167378e-05, - "loss": 0.1552, + "epoch": 3.09, + "learning_rate": 3.839122486288849e-05, + "loss": 0.1452, "step": 1690 }, { - "epoch": 2.42, - "learning_rate": 4.212106843685791e-05, - "loss": 0.1657, + "epoch": 3.11, + "learning_rate": 3.828966077594963e-05, + "loss": 0.1661, "step": 1700 }, { - "epoch": 2.43, - "learning_rate": 4.204204204204204e-05, - "loss": 0.1696, + "epoch": 3.13, + "learning_rate": 3.818809668901077e-05, + "loss": 0.1367, "step": 1710 }, { - "epoch": 2.45, - "learning_rate": 4.1963015647226175e-05, - "loss": 0.1567, + "epoch": 3.14, + "learning_rate": 3.808653260207191e-05, + "loss": 0.1345, "step": 1720 }, { - "epoch": 2.46, - "learning_rate": 4.1883989252410307e-05, - "loss": 0.1518, + "epoch": 3.16, + "learning_rate": 3.798496851513305e-05, + "loss": 0.1143, "step": 1730 }, { - "epoch": 2.47, - "learning_rate": 4.180496285759444e-05, - "loss": 0.1744, + "epoch": 3.18, + "learning_rate": 3.788340442819419e-05, + "loss": 0.1928, "step": 1740 }, { - "epoch": 2.49, - "learning_rate": 4.172593646277857e-05, - "loss": 0.1696, + "epoch": 3.2, + "learning_rate": 3.778184034125533e-05, + "loss": 0.152, "step": 1750 }, { - "epoch": 2.5, - "learning_rate": 4.16469100679627e-05, - "loss": 0.1637, + "epoch": 3.22, + "learning_rate": 3.768027625431648e-05, + "loss": 0.1454, "step": 1760 }, { - "epoch": 2.52, - "learning_rate": 4.156788367314683e-05, - "loss": 0.147, + "epoch": 3.24, + "learning_rate": 3.757871216737762e-05, + "loss": 0.1523, "step": 1770 }, { - "epoch": 2.53, - "learning_rate": 4.1488857278330965e-05, - "loss": 0.1602, + "epoch": 3.25, + "learning_rate": 3.7477148080438754e-05, + "loss": 0.1212, "step": 1780 }, { - "epoch": 2.55, - "learning_rate": 4.14098308835151e-05, - "loss": 0.1275, + "epoch": 3.27, + "learning_rate": 3.73755839934999e-05, + "loss": 0.159, "step": 1790 }, { - "epoch": 2.56, - "learning_rate": 4.133080448869923e-05, - "loss": 0.1472, + "epoch": 3.29, + "learning_rate": 3.727401990656104e-05, + "loss": 0.1403, "step": 1800 }, { - "epoch": 2.57, - "learning_rate": 4.125177809388336e-05, - "loss": 0.1812, + "epoch": 3.31, + "learning_rate": 3.717245581962218e-05, + "loss": 0.1345, "step": 1810 }, { - "epoch": 2.59, - "learning_rate": 4.117275169906749e-05, - "loss": 0.1632, + "epoch": 3.33, + "learning_rate": 3.707089173268333e-05, + "loss": 0.1669, "step": 1820 }, { - "epoch": 2.6, - "learning_rate": 4.1093725304251623e-05, - "loss": 0.169, + "epoch": 3.35, + "learning_rate": 3.6969327645744464e-05, + "loss": 0.1526, "step": 1830 }, { - "epoch": 2.62, - "learning_rate": 4.1014698909435755e-05, - "loss": 0.1457, + "epoch": 3.36, + "learning_rate": 3.6867763558805604e-05, + "loss": 0.1423, "step": 1840 }, { - "epoch": 2.63, - "learning_rate": 4.093567251461988e-05, - "loss": 0.1599, + "epoch": 3.38, + "learning_rate": 3.676619947186675e-05, + "loss": 0.1443, "step": 1850 }, { - "epoch": 2.64, - "learning_rate": 4.085664611980402e-05, - "loss": 0.1672, + "epoch": 3.4, + "learning_rate": 3.666463538492789e-05, + "loss": 0.1467, "step": 1860 }, { - "epoch": 2.66, - "learning_rate": 4.077761972498815e-05, - "loss": 0.1473, + "epoch": 3.42, + "learning_rate": 3.656307129798903e-05, + "loss": 0.146, "step": 1870 }, { - "epoch": 2.67, - "learning_rate": 4.069859333017228e-05, - "loss": 0.1584, + "epoch": 3.44, + "learning_rate": 3.6461507211050174e-05, + "loss": 0.1397, "step": 1880 }, { - "epoch": 2.69, - "learning_rate": 4.0619566935356414e-05, - "loss": 0.1474, + "epoch": 3.46, + "learning_rate": 3.6359943124111314e-05, + "loss": 0.1533, "step": 1890 }, { - "epoch": 2.7, - "learning_rate": 4.0540540540540545e-05, - "loss": 0.1658, + "epoch": 3.47, + "learning_rate": 3.6258379037172455e-05, + "loss": 0.1379, "step": 1900 }, { - "epoch": 2.72, - "learning_rate": 4.046151414572467e-05, - "loss": 0.156, + "epoch": 3.49, + "learning_rate": 3.6156814950233596e-05, + "loss": 0.1572, "step": 1910 }, { - "epoch": 2.73, - "learning_rate": 4.03824877509088e-05, - "loss": 0.1563, + "epoch": 3.51, + "learning_rate": 3.605525086329474e-05, + "loss": 0.1483, "step": 1920 }, { - "epoch": 2.74, - "learning_rate": 4.0303461356092934e-05, - "loss": 0.1382, + "epoch": 3.53, + "learning_rate": 3.5953686776355884e-05, + "loss": 0.1603, "step": 1930 }, { - "epoch": 2.76, - "learning_rate": 4.022443496127707e-05, - "loss": 0.1536, + "epoch": 3.55, + "learning_rate": 3.5852122689417025e-05, + "loss": 0.1434, "step": 1940 }, { - "epoch": 2.77, - "learning_rate": 4.0145408566461204e-05, - "loss": 0.1579, + "epoch": 3.56, + "learning_rate": 3.5750558602478165e-05, + "loss": 0.1655, "step": 1950 }, { - "epoch": 2.79, - "learning_rate": 4.0066382171645335e-05, - "loss": 0.1245, + "epoch": 3.58, + "learning_rate": 3.5648994515539306e-05, + "loss": 0.1564, "step": 1960 }, { - "epoch": 2.8, - "learning_rate": 3.998735577682946e-05, - "loss": 0.1648, + "epoch": 3.6, + "learning_rate": 3.554743042860045e-05, + "loss": 0.1335, "step": 1970 }, { - "epoch": 2.82, - "learning_rate": 3.990832938201359e-05, - "loss": 0.1703, + "epoch": 3.62, + "learning_rate": 3.544586634166159e-05, + "loss": 0.1419, "step": 1980 }, { - "epoch": 2.83, - "learning_rate": 3.9829302987197724e-05, - "loss": 0.1521, + "epoch": 3.64, + "learning_rate": 3.5344302254722735e-05, + "loss": 0.1464, "step": 1990 }, { - "epoch": 2.84, - "learning_rate": 3.9750276592381855e-05, - "loss": 0.1543, + "epoch": 3.66, + "learning_rate": 3.5242738167783875e-05, + "loss": 0.1423, "step": 2000 }, { - "epoch": 2.86, - "learning_rate": 3.967125019756599e-05, - "loss": 0.1588, + "epoch": 3.67, + "learning_rate": 3.5141174080845016e-05, + "loss": 0.1425, "step": 2010 }, { - "epoch": 2.87, - "learning_rate": 3.9592223802750125e-05, - "loss": 0.1234, + "epoch": 3.69, + "learning_rate": 3.503960999390616e-05, + "loss": 0.1429, "step": 2020 }, { - "epoch": 2.89, - "learning_rate": 3.951319740793425e-05, - "loss": 0.1407, + "epoch": 3.71, + "learning_rate": 3.49380459069673e-05, + "loss": 0.1237, "step": 2030 }, { - "epoch": 2.9, - "learning_rate": 3.943417101311838e-05, - "loss": 0.1547, + "epoch": 3.73, + "learning_rate": 3.483648182002844e-05, + "loss": 0.1571, "step": 2040 }, { - "epoch": 2.92, - "learning_rate": 3.9355144618302514e-05, - "loss": 0.1271, + "epoch": 3.75, + "learning_rate": 3.4734917733089586e-05, + "loss": 0.1504, "step": 2050 }, { - "epoch": 2.93, - "learning_rate": 3.9276118223486645e-05, - "loss": 0.1231, + "epoch": 3.77, + "learning_rate": 3.4633353646150726e-05, + "loss": 0.1423, "step": 2060 }, { - "epoch": 2.94, - "learning_rate": 3.919709182867078e-05, - "loss": 0.1596, + "epoch": 3.78, + "learning_rate": 3.453178955921186e-05, + "loss": 0.1373, "step": 2070 }, { - "epoch": 2.96, - "learning_rate": 3.911806543385491e-05, - "loss": 0.1723, + "epoch": 3.8, + "learning_rate": 3.4430225472273e-05, + "loss": 0.1387, "step": 2080 }, { - "epoch": 2.97, - "learning_rate": 3.903903903903904e-05, - "loss": 0.1504, + "epoch": 3.82, + "learning_rate": 3.432866138533415e-05, + "loss": 0.1628, "step": 2090 }, { - "epoch": 2.99, - "learning_rate": 3.896001264422317e-05, - "loss": 0.1427, + "epoch": 3.84, + "learning_rate": 3.422709729839529e-05, + "loss": 0.134, "step": 2100 }, { - "epoch": 3.0, - "eval_accuracy": 0.9761, - "eval_loss": 0.06038949638605118, - "eval_runtime": 59.8642, - "eval_samples_per_second": 167.045, - "eval_steps_per_second": 5.229, - "step": 2109 - }, - { - "epoch": 3.0, - "learning_rate": 3.8880986249407304e-05, - "loss": 0.1521, + "epoch": 3.86, + "learning_rate": 3.412553321145643e-05, + "loss": 0.1636, "step": 2110 }, { - "epoch": 3.01, - "learning_rate": 3.8801959854591436e-05, - "loss": 0.1344, + "epoch": 3.88, + "learning_rate": 3.402396912451758e-05, + "loss": 0.1548, "step": 2120 }, { - "epoch": 3.03, - "learning_rate": 3.872293345977557e-05, - "loss": 0.133, + "epoch": 3.89, + "learning_rate": 3.392240503757871e-05, + "loss": 0.1643, "step": 2130 }, { - "epoch": 3.04, - "learning_rate": 3.86439070649597e-05, - "loss": 0.141, + "epoch": 3.91, + "learning_rate": 3.382084095063985e-05, + "loss": 0.1427, "step": 2140 }, { - "epoch": 3.06, - "learning_rate": 3.856488067014383e-05, - "loss": 0.1306, + "epoch": 3.93, + "learning_rate": 3.3719276863701e-05, + "loss": 0.1171, "step": 2150 }, { - "epoch": 3.07, - "learning_rate": 3.848585427532796e-05, - "loss": 0.1694, + "epoch": 3.95, + "learning_rate": 3.361771277676214e-05, + "loss": 0.1536, "step": 2160 }, { - "epoch": 3.09, - "learning_rate": 3.8406827880512094e-05, - "loss": 0.1679, + "epoch": 3.97, + "learning_rate": 3.351614868982328e-05, + "loss": 0.1432, "step": 2170 }, { - "epoch": 3.1, - "learning_rate": 3.8327801485696226e-05, - "loss": 0.1388, + "epoch": 3.99, + "learning_rate": 3.341458460288442e-05, + "loss": 0.1424, "step": 2180 }, { - "epoch": 3.11, - "learning_rate": 3.824877509088036e-05, - "loss": 0.1369, + "epoch": 4.0, + "eval_accuracy": 0.9686333333333333, + "eval_loss": 0.08514226227998734, + "eval_runtime": 107.4467, + "eval_samples_per_second": 279.208, + "eval_steps_per_second": 8.73, + "step": 2188 + }, + { + "epoch": 4.0, + "learning_rate": 3.331302051594556e-05, + "loss": 0.1166, "step": 2190 }, { - "epoch": 3.13, - "learning_rate": 3.816974869606449e-05, - "loss": 0.1535, + "epoch": 4.02, + "learning_rate": 3.32114564290067e-05, + "loss": 0.1505, "step": 2200 }, { - "epoch": 3.14, - "learning_rate": 3.809072230124862e-05, - "loss": 0.1499, + "epoch": 4.04, + "learning_rate": 3.310989234206784e-05, + "loss": 0.125, "step": 2210 }, { - "epoch": 3.16, - "learning_rate": 3.8011695906432746e-05, - "loss": 0.1609, + "epoch": 4.06, + "learning_rate": 3.300832825512899e-05, + "loss": 0.1363, "step": 2220 }, { - "epoch": 3.17, - "learning_rate": 3.793266951161688e-05, - "loss": 0.1402, + "epoch": 4.08, + "learning_rate": 3.290676416819013e-05, + "loss": 0.1539, "step": 2230 }, { - "epoch": 3.19, - "learning_rate": 3.7853643116801016e-05, - "loss": 0.1305, + "epoch": 4.1, + "learning_rate": 3.280520008125127e-05, + "loss": 0.1323, "step": 2240 }, { - "epoch": 3.2, - "learning_rate": 3.777461672198515e-05, - "loss": 0.1483, + "epoch": 4.11, + "learning_rate": 3.270363599431241e-05, + "loss": 0.1549, "step": 2250 }, { - "epoch": 3.21, - "learning_rate": 3.769559032716928e-05, - "loss": 0.1322, + "epoch": 4.13, + "learning_rate": 3.260207190737355e-05, + "loss": 0.1708, "step": 2260 }, { - "epoch": 3.23, - "learning_rate": 3.761656393235341e-05, - "loss": 0.134, + "epoch": 4.15, + "learning_rate": 3.2500507820434694e-05, + "loss": 0.1367, "step": 2270 }, { - "epoch": 3.24, - "learning_rate": 3.7537537537537536e-05, - "loss": 0.1399, + "epoch": 4.17, + "learning_rate": 3.239894373349584e-05, + "loss": 0.1474, "step": 2280 }, { - "epoch": 3.26, - "learning_rate": 3.745851114272167e-05, - "loss": 0.1505, + "epoch": 4.19, + "learning_rate": 3.229737964655698e-05, + "loss": 0.1296, "step": 2290 }, { - "epoch": 3.27, - "learning_rate": 3.73794847479058e-05, - "loss": 0.1433, + "epoch": 4.2, + "learning_rate": 3.219581555961812e-05, + "loss": 0.1349, "step": 2300 }, { - "epoch": 3.28, - "learning_rate": 3.730045835308993e-05, - "loss": 0.1272, + "epoch": 4.22, + "learning_rate": 3.2094251472679257e-05, + "loss": 0.1239, "step": 2310 }, { - "epoch": 3.3, - "learning_rate": 3.722143195827407e-05, - "loss": 0.1498, + "epoch": 4.24, + "learning_rate": 3.1992687385740404e-05, + "loss": 0.1555, "step": 2320 }, { - "epoch": 3.31, - "learning_rate": 3.71424055634582e-05, - "loss": 0.1288, + "epoch": 4.26, + "learning_rate": 3.1891123298801545e-05, + "loss": 0.1417, "step": 2330 }, { - "epoch": 3.33, - "learning_rate": 3.7063379168642326e-05, - "loss": 0.1553, + "epoch": 4.28, + "learning_rate": 3.1789559211862685e-05, + "loss": 0.1362, "step": 2340 }, { - "epoch": 3.34, - "learning_rate": 3.698435277382646e-05, - "loss": 0.1656, + "epoch": 4.3, + "learning_rate": 3.168799512492383e-05, + "loss": 0.1335, "step": 2350 }, { - "epoch": 3.36, - "learning_rate": 3.690532637901059e-05, - "loss": 0.1336, + "epoch": 4.31, + "learning_rate": 3.1586431037984973e-05, + "loss": 0.145, "step": 2360 }, { - "epoch": 3.37, - "learning_rate": 3.682629998419472e-05, - "loss": 0.1469, + "epoch": 4.33, + "learning_rate": 3.148486695104611e-05, + "loss": 0.1235, "step": 2370 }, { - "epoch": 3.38, - "learning_rate": 3.674727358937885e-05, - "loss": 0.1254, + "epoch": 4.35, + "learning_rate": 3.1383302864107255e-05, + "loss": 0.1101, "step": 2380 }, { - "epoch": 3.4, - "learning_rate": 3.666824719456299e-05, - "loss": 0.1366, + "epoch": 4.37, + "learning_rate": 3.1281738777168395e-05, + "loss": 0.1134, "step": 2390 }, { - "epoch": 3.41, - "learning_rate": 3.6589220799747116e-05, - "loss": 0.1333, + "epoch": 4.39, + "learning_rate": 3.1180174690229536e-05, + "loss": 0.1345, "step": 2400 }, { - "epoch": 3.43, - "learning_rate": 3.651019440493125e-05, - "loss": 0.1435, + "epoch": 4.41, + "learning_rate": 3.107861060329068e-05, + "loss": 0.1466, "step": 2410 }, { - "epoch": 3.44, - "learning_rate": 3.643116801011538e-05, - "loss": 0.1488, + "epoch": 4.42, + "learning_rate": 3.0977046516351824e-05, + "loss": 0.1239, "step": 2420 }, { - "epoch": 3.46, - "learning_rate": 3.635214161529951e-05, - "loss": 0.1422, + "epoch": 4.44, + "learning_rate": 3.087548242941296e-05, + "loss": 0.1315, "step": 2430 }, { - "epoch": 3.47, - "learning_rate": 3.627311522048364e-05, - "loss": 0.1267, + "epoch": 4.46, + "learning_rate": 3.07739183424741e-05, + "loss": 0.1165, "step": 2440 }, { - "epoch": 3.48, - "learning_rate": 3.6194088825667775e-05, - "loss": 0.1257, + "epoch": 4.48, + "learning_rate": 3.0672354255535246e-05, + "loss": 0.1401, "step": 2450 }, { - "epoch": 3.5, - "learning_rate": 3.6115062430851906e-05, - "loss": 0.1362, + "epoch": 4.5, + "learning_rate": 3.057079016859639e-05, + "loss": 0.1428, "step": 2460 }, { - "epoch": 3.51, - "learning_rate": 3.603603603603604e-05, - "loss": 0.1263, + "epoch": 4.52, + "learning_rate": 3.0469226081657524e-05, + "loss": 0.1198, "step": 2470 }, { - "epoch": 3.53, - "learning_rate": 3.595700964122017e-05, - "loss": 0.1317, + "epoch": 4.53, + "learning_rate": 3.036766199471867e-05, + "loss": 0.1192, "step": 2480 }, { - "epoch": 3.54, - "learning_rate": 3.58779832464043e-05, - "loss": 0.1268, + "epoch": 4.55, + "learning_rate": 3.0266097907779812e-05, + "loss": 0.1212, "step": 2490 }, { - "epoch": 3.55, - "learning_rate": 3.579895685158843e-05, - "loss": 0.156, + "epoch": 4.57, + "learning_rate": 3.016453382084095e-05, + "loss": 0.1148, "step": 2500 }, { - "epoch": 3.57, - "learning_rate": 3.5719930456772565e-05, - "loss": 0.1366, + "epoch": 4.59, + "learning_rate": 3.006296973390209e-05, + "loss": 0.1472, "step": 2510 }, { - "epoch": 3.58, - "learning_rate": 3.564090406195669e-05, - "loss": 0.1442, + "epoch": 4.61, + "learning_rate": 2.9961405646963238e-05, + "loss": 0.1403, "step": 2520 }, { - "epoch": 3.6, - "learning_rate": 3.556187766714083e-05, - "loss": 0.1228, + "epoch": 4.63, + "learning_rate": 2.9859841560024375e-05, + "loss": 0.1353, "step": 2530 }, { - "epoch": 3.61, - "learning_rate": 3.548285127232496e-05, - "loss": 0.1321, + "epoch": 4.64, + "learning_rate": 2.9758277473085516e-05, + "loss": 0.15, "step": 2540 }, { - "epoch": 3.63, - "learning_rate": 3.540382487750909e-05, - "loss": 0.1358, + "epoch": 4.66, + "learning_rate": 2.9656713386146663e-05, + "loss": 0.1106, "step": 2550 }, { - "epoch": 3.64, - "learning_rate": 3.532479848269322e-05, - "loss": 0.1415, + "epoch": 4.68, + "learning_rate": 2.95551492992078e-05, + "loss": 0.1476, "step": 2560 }, { - "epoch": 3.65, - "learning_rate": 3.5245772087877355e-05, - "loss": 0.1324, + "epoch": 4.7, + "learning_rate": 2.945358521226894e-05, + "loss": 0.141, "step": 2570 }, { - "epoch": 3.67, - "learning_rate": 3.5166745693061487e-05, + "epoch": 4.72, + "learning_rate": 2.935202112533009e-05, "loss": 0.1299, "step": 2580 }, { - "epoch": 3.68, - "learning_rate": 3.508771929824561e-05, - "loss": 0.1263, + "epoch": 4.73, + "learning_rate": 2.9250457038391226e-05, + "loss": 0.1546, "step": 2590 }, { - "epoch": 3.7, - "learning_rate": 3.500869290342974e-05, - "loss": 0.1134, + "epoch": 4.75, + "learning_rate": 2.9148892951452366e-05, + "loss": 0.175, "step": 2600 }, { - "epoch": 3.71, - "learning_rate": 3.492966650861388e-05, - "loss": 0.1313, + "epoch": 4.77, + "learning_rate": 2.9047328864513507e-05, + "loss": 0.1388, "step": 2610 }, { - "epoch": 3.73, - "learning_rate": 3.485064011379801e-05, - "loss": 0.1138, + "epoch": 4.79, + "learning_rate": 2.894576477757465e-05, + "loss": 0.1279, "step": 2620 }, { - "epoch": 3.74, - "learning_rate": 3.4771613718982145e-05, - "loss": 0.174, + "epoch": 4.81, + "learning_rate": 2.8844200690635792e-05, + "loss": 0.135, "step": 2630 }, { - "epoch": 3.75, - "learning_rate": 3.469258732416628e-05, - "loss": 0.1235, + "epoch": 4.83, + "learning_rate": 2.8742636603696933e-05, + "loss": 0.101, "step": 2640 }, { - "epoch": 3.77, - "learning_rate": 3.46135609293504e-05, - "loss": 0.1267, + "epoch": 4.84, + "learning_rate": 2.8641072516758077e-05, + "loss": 0.1344, "step": 2650 }, { - "epoch": 3.78, - "learning_rate": 3.453453453453453e-05, - "loss": 0.1578, + "epoch": 4.86, + "learning_rate": 2.8539508429819217e-05, + "loss": 0.1122, "step": 2660 }, { - "epoch": 3.8, - "learning_rate": 3.4455508139718665e-05, - "loss": 0.1508, + "epoch": 4.88, + "learning_rate": 2.8437944342880358e-05, + "loss": 0.1043, "step": 2670 }, { - "epoch": 3.81, - "learning_rate": 3.43764817449028e-05, - "loss": 0.1189, + "epoch": 4.9, + "learning_rate": 2.8336380255941502e-05, + "loss": 0.1296, "step": 2680 }, { - "epoch": 3.83, - "learning_rate": 3.4297455350086935e-05, - "loss": 0.1456, + "epoch": 4.92, + "learning_rate": 2.8234816169002643e-05, + "loss": 0.1195, "step": 2690 }, { - "epoch": 3.84, - "learning_rate": 3.421842895527107e-05, - "loss": 0.1214, + "epoch": 4.94, + "learning_rate": 2.8133252082063783e-05, + "loss": 0.117, "step": 2700 }, { - "epoch": 3.85, - "learning_rate": 3.413940256045519e-05, - "loss": 0.1493, + "epoch": 4.95, + "learning_rate": 2.8031687995124927e-05, + "loss": 0.1098, "step": 2710 }, { - "epoch": 3.87, - "learning_rate": 3.4060376165639323e-05, - "loss": 0.1317, + "epoch": 4.97, + "learning_rate": 2.7930123908186068e-05, + "loss": 0.1168, "step": 2720 }, { - "epoch": 3.88, - "learning_rate": 3.3981349770823455e-05, - "loss": 0.133, + "epoch": 4.99, + "learning_rate": 2.782855982124721e-05, + "loss": 0.1244, "step": 2730 }, { - "epoch": 3.9, - "learning_rate": 3.390232337600759e-05, - "loss": 0.1219, + "epoch": 5.0, + "eval_accuracy": 0.9732666666666666, + "eval_loss": 0.07139687240123749, + "eval_runtime": 107.2208, + "eval_samples_per_second": 279.796, + "eval_steps_per_second": 8.748, + "step": 2735 + }, + { + "epoch": 5.01, + "learning_rate": 2.7726995734308346e-05, + "loss": 0.1521, "step": 2740 }, { - "epoch": 3.91, - "learning_rate": 3.382329698119172e-05, - "loss": 0.1372, + "epoch": 5.03, + "learning_rate": 2.7625431647369494e-05, + "loss": 0.1246, "step": 2750 }, { - "epoch": 3.92, - "learning_rate": 3.374427058637585e-05, - "loss": 0.1372, + "epoch": 5.05, + "learning_rate": 2.7523867560430634e-05, + "loss": 0.1046, "step": 2760 }, { - "epoch": 3.94, - "learning_rate": 3.366524419155998e-05, - "loss": 0.1458, + "epoch": 5.06, + "learning_rate": 2.742230347349177e-05, + "loss": 0.1186, "step": 2770 }, { - "epoch": 3.95, - "learning_rate": 3.3586217796744114e-05, - "loss": 0.1455, + "epoch": 5.08, + "learning_rate": 2.732073938655292e-05, + "loss": 0.1086, "step": 2780 }, { - "epoch": 3.97, - "learning_rate": 3.3507191401928245e-05, - "loss": 0.1508, + "epoch": 5.1, + "learning_rate": 2.721917529961406e-05, + "loss": 0.1158, "step": 2790 }, { - "epoch": 3.98, - "learning_rate": 3.342816500711238e-05, - "loss": 0.1458, + "epoch": 5.12, + "learning_rate": 2.7117611212675197e-05, + "loss": 0.1199, "step": 2800 }, { - "epoch": 4.0, - "learning_rate": 3.334913861229651e-05, - "loss": 0.143, + "epoch": 5.14, + "learning_rate": 2.7016047125736344e-05, + "loss": 0.1027, "step": 2810 }, { - "epoch": 4.0, - "eval_accuracy": 0.97, - "eval_loss": 0.08071798086166382, - "eval_runtime": 59.8831, - "eval_samples_per_second": 166.992, - "eval_steps_per_second": 5.227, - "step": 2813 - }, - { - "epoch": 4.01, - "learning_rate": 3.327011221748064e-05, - "loss": 0.1265, + "epoch": 5.16, + "learning_rate": 2.6914483038797485e-05, + "loss": 0.1379, "step": 2820 }, { - "epoch": 4.02, - "learning_rate": 3.319108582266477e-05, - "loss": 0.129, + "epoch": 5.17, + "learning_rate": 2.6812918951858622e-05, + "loss": 0.1285, "step": 2830 }, { - "epoch": 4.04, - "learning_rate": 3.3112059427848904e-05, - "loss": 0.1364, + "epoch": 5.19, + "learning_rate": 2.6711354864919763e-05, + "loss": 0.1068, "step": 2840 }, { - "epoch": 4.05, - "learning_rate": 3.3033033033033035e-05, - "loss": 0.1148, + "epoch": 5.21, + "learning_rate": 2.6609790777980907e-05, + "loss": 0.1201, "step": 2850 }, { - "epoch": 4.07, - "learning_rate": 3.295400663821717e-05, - "loss": 0.1114, + "epoch": 5.23, + "learning_rate": 2.6508226691042048e-05, + "loss": 0.1081, "step": 2860 }, { - "epoch": 4.08, - "learning_rate": 3.28749802434013e-05, - "loss": 0.1081, + "epoch": 5.25, + "learning_rate": 2.640666260410319e-05, + "loss": 0.1378, "step": 2870 }, { - "epoch": 4.1, - "learning_rate": 3.279595384858543e-05, - "loss": 0.138, + "epoch": 5.27, + "learning_rate": 2.6305098517164332e-05, + "loss": 0.1165, "step": 2880 }, { - "epoch": 4.11, - "learning_rate": 3.2716927453769555e-05, - "loss": 0.1163, + "epoch": 5.28, + "learning_rate": 2.6203534430225473e-05, + "loss": 0.0983, "step": 2890 }, { - "epoch": 4.12, - "learning_rate": 3.2637901058953694e-05, - "loss": 0.1348, + "epoch": 5.3, + "learning_rate": 2.6101970343286614e-05, + "loss": 0.1398, "step": 2900 }, { - "epoch": 4.14, - "learning_rate": 3.2558874664137825e-05, - "loss": 0.1249, + "epoch": 5.32, + "learning_rate": 2.6000406256347758e-05, + "loss": 0.0956, "step": 2910 }, { - "epoch": 4.15, - "learning_rate": 3.247984826932196e-05, - "loss": 0.1094, + "epoch": 5.34, + "learning_rate": 2.58988421694089e-05, + "loss": 0.1129, "step": 2920 }, { - "epoch": 4.17, - "learning_rate": 3.240082187450609e-05, - "loss": 0.1124, + "epoch": 5.36, + "learning_rate": 2.579727808247004e-05, + "loss": 0.1347, "step": 2930 }, { - "epoch": 4.18, - "learning_rate": 3.232179547969022e-05, - "loss": 0.1181, + "epoch": 5.37, + "learning_rate": 2.569571399553118e-05, + "loss": 0.1196, "step": 2940 }, { - "epoch": 4.19, - "learning_rate": 3.2242769084874345e-05, - "loss": 0.1161, + "epoch": 5.39, + "learning_rate": 2.5594149908592324e-05, + "loss": 0.119, "step": 2950 }, { - "epoch": 4.21, - "learning_rate": 3.216374269005848e-05, - "loss": 0.1365, + "epoch": 5.41, + "learning_rate": 2.5492585821653465e-05, + "loss": 0.1259, "step": 2960 }, { - "epoch": 4.22, - "learning_rate": 3.208471629524261e-05, - "loss": 0.1183, + "epoch": 5.43, + "learning_rate": 2.5391021734714605e-05, + "loss": 0.1212, "step": 2970 }, { - "epoch": 4.24, - "learning_rate": 3.200568990042675e-05, - "loss": 0.1288, + "epoch": 5.45, + "learning_rate": 2.528945764777575e-05, + "loss": 0.1481, "step": 2980 }, { - "epoch": 4.25, - "learning_rate": 3.192666350561088e-05, - "loss": 0.1042, + "epoch": 5.47, + "learning_rate": 2.518789356083689e-05, + "loss": 0.0944, "step": 2990 }, { - "epoch": 4.27, - "learning_rate": 3.184763711079501e-05, - "loss": 0.1182, + "epoch": 5.48, + "learning_rate": 2.508632947389803e-05, + "loss": 0.1074, "step": 3000 }, { - "epoch": 4.28, - "learning_rate": 3.176861071597914e-05, - "loss": 0.133, + "epoch": 5.5, + "learning_rate": 2.498476538695917e-05, + "loss": 0.1107, "step": 3010 }, { - "epoch": 4.29, - "learning_rate": 3.168958432116327e-05, - "loss": 0.143, + "epoch": 5.52, + "learning_rate": 2.4883201300020315e-05, + "loss": 0.1103, "step": 3020 }, { - "epoch": 4.31, - "learning_rate": 3.16105579263474e-05, - "loss": 0.1564, + "epoch": 5.54, + "learning_rate": 2.4781637213081456e-05, + "loss": 0.128, "step": 3030 }, { - "epoch": 4.32, - "learning_rate": 3.153153153153153e-05, - "loss": 0.1307, + "epoch": 5.56, + "learning_rate": 2.4680073126142597e-05, + "loss": 0.0964, "step": 3040 }, { - "epoch": 4.34, - "learning_rate": 3.145250513671566e-05, - "loss": 0.1084, + "epoch": 5.58, + "learning_rate": 2.457850903920374e-05, + "loss": 0.1367, "step": 3050 }, { - "epoch": 4.35, - "learning_rate": 3.13734787418998e-05, - "loss": 0.1435, + "epoch": 5.59, + "learning_rate": 2.4476944952264878e-05, + "loss": 0.1317, "step": 3060 }, { - "epoch": 4.37, - "learning_rate": 3.129445234708393e-05, - "loss": 0.1394, + "epoch": 5.61, + "learning_rate": 2.4375380865326022e-05, + "loss": 0.116, "step": 3070 }, { - "epoch": 4.38, - "learning_rate": 3.121542595226806e-05, - "loss": 0.1206, + "epoch": 5.63, + "learning_rate": 2.4273816778387163e-05, + "loss": 0.1198, "step": 3080 }, { - "epoch": 4.39, - "learning_rate": 3.113639955745219e-05, - "loss": 0.1361, + "epoch": 5.65, + "learning_rate": 2.4172252691448303e-05, + "loss": 0.1302, "step": 3090 }, { - "epoch": 4.41, - "learning_rate": 3.105737316263632e-05, - "loss": 0.109, + "epoch": 5.67, + "learning_rate": 2.4070688604509447e-05, + "loss": 0.1217, "step": 3100 }, { - "epoch": 4.42, - "learning_rate": 3.097834676782045e-05, - "loss": 0.0935, + "epoch": 5.69, + "learning_rate": 2.3969124517570588e-05, + "loss": 0.1108, "step": 3110 }, { - "epoch": 4.44, - "learning_rate": 3.0899320373004584e-05, - "loss": 0.1217, + "epoch": 5.7, + "learning_rate": 2.386756043063173e-05, + "loss": 0.1093, "step": 3120 }, { - "epoch": 4.45, - "learning_rate": 3.0820293978188716e-05, - "loss": 0.1139, + "epoch": 5.72, + "learning_rate": 2.376599634369287e-05, + "loss": 0.1178, "step": 3130 }, { - "epoch": 4.46, - "learning_rate": 3.074126758337285e-05, - "loss": 0.1201, + "epoch": 5.74, + "learning_rate": 2.3664432256754014e-05, + "loss": 0.1315, "step": 3140 }, { - "epoch": 4.48, - "learning_rate": 3.066224118855698e-05, - "loss": 0.1222, + "epoch": 5.76, + "learning_rate": 2.3562868169815154e-05, + "loss": 0.1138, "step": 3150 }, { - "epoch": 4.49, - "learning_rate": 3.058321479374111e-05, - "loss": 0.1296, + "epoch": 5.78, + "learning_rate": 2.3461304082876295e-05, + "loss": 0.1099, "step": 3160 }, { - "epoch": 4.51, - "learning_rate": 3.0504188398925243e-05, - "loss": 0.1308, + "epoch": 5.8, + "learning_rate": 2.335973999593744e-05, + "loss": 0.1269, "step": 3170 }, { - "epoch": 4.52, - "learning_rate": 3.0425162004109374e-05, - "loss": 0.0969, + "epoch": 5.81, + "learning_rate": 2.3258175908998576e-05, + "loss": 0.1086, "step": 3180 }, { - "epoch": 4.54, - "learning_rate": 3.034613560929351e-05, - "loss": 0.1037, + "epoch": 5.83, + "learning_rate": 2.315661182205972e-05, + "loss": 0.1093, "step": 3190 }, { - "epoch": 4.55, - "learning_rate": 3.0267109214477634e-05, - "loss": 0.1152, + "epoch": 5.85, + "learning_rate": 2.3055047735120864e-05, + "loss": 0.1157, "step": 3200 }, { - "epoch": 4.56, - "learning_rate": 3.0188082819661766e-05, - "loss": 0.1205, + "epoch": 5.87, + "learning_rate": 2.2953483648182e-05, + "loss": 0.1169, "step": 3210 }, { - "epoch": 4.58, - "learning_rate": 3.01090564248459e-05, - "loss": 0.115, + "epoch": 5.89, + "learning_rate": 2.2851919561243146e-05, + "loss": 0.1128, "step": 3220 }, { - "epoch": 4.59, - "learning_rate": 3.0030030030030033e-05, - "loss": 0.1311, + "epoch": 5.9, + "learning_rate": 2.2750355474304286e-05, + "loss": 0.1142, "step": 3230 }, { - "epoch": 4.61, - "learning_rate": 2.9951003635214164e-05, - "loss": 0.1366, + "epoch": 5.92, + "learning_rate": 2.2648791387365427e-05, + "loss": 0.1058, "step": 3240 }, { - "epoch": 4.62, - "learning_rate": 2.9871977240398296e-05, - "loss": 0.1323, + "epoch": 5.94, + "learning_rate": 2.254722730042657e-05, + "loss": 0.1145, "step": 3250 }, { - "epoch": 4.64, - "learning_rate": 2.9792950845582424e-05, - "loss": 0.1266, + "epoch": 5.96, + "learning_rate": 2.2445663213487712e-05, + "loss": 0.1027, "step": 3260 }, { - "epoch": 4.65, - "learning_rate": 2.9713924450766556e-05, - "loss": 0.1274, + "epoch": 5.98, + "learning_rate": 2.2344099126548852e-05, + "loss": 0.1139, "step": 3270 }, { - "epoch": 4.66, - "learning_rate": 2.9634898055950688e-05, - "loss": 0.1481, + "epoch": 6.0, + "learning_rate": 2.2242535039609997e-05, + "loss": 0.1089, "step": 3280 }, { - "epoch": 4.68, - "learning_rate": 2.955587166113482e-05, - "loss": 0.1298, + "epoch": 6.0, + "eval_accuracy": 0.9733666666666667, + "eval_loss": 0.07119759172201157, + "eval_runtime": 106.9967, + "eval_samples_per_second": 280.382, + "eval_steps_per_second": 8.767, + "step": 3282 + }, + { + "epoch": 6.01, + "learning_rate": 2.2140970952671137e-05, + "loss": 0.096, "step": 3290 }, { - "epoch": 4.69, - "learning_rate": 2.9476845266318955e-05, - "loss": 0.1301, + "epoch": 6.03, + "learning_rate": 2.2039406865732278e-05, + "loss": 0.1201, "step": 3300 }, { - "epoch": 4.71, - "learning_rate": 2.9397818871503086e-05, - "loss": 0.1176, + "epoch": 6.05, + "learning_rate": 2.193784277879342e-05, + "loss": 0.1199, "step": 3310 }, { - "epoch": 4.72, - "learning_rate": 2.931879247668721e-05, - "loss": 0.1169, + "epoch": 6.07, + "learning_rate": 2.1836278691854563e-05, + "loss": 0.1293, "step": 3320 }, { - "epoch": 4.74, - "learning_rate": 2.9239766081871346e-05, - "loss": 0.1274, + "epoch": 6.09, + "learning_rate": 2.1734714604915703e-05, + "loss": 0.1067, "step": 3330 }, { - "epoch": 4.75, - "learning_rate": 2.9160739687055478e-05, - "loss": 0.1128, + "epoch": 6.11, + "learning_rate": 2.1633150517976844e-05, + "loss": 0.1032, "step": 3340 }, { - "epoch": 4.76, - "learning_rate": 2.908171329223961e-05, - "loss": 0.1389, + "epoch": 6.12, + "learning_rate": 2.1531586431037988e-05, + "loss": 0.1162, "step": 3350 }, { - "epoch": 4.78, - "learning_rate": 2.900268689742374e-05, - "loss": 0.1013, + "epoch": 6.14, + "learning_rate": 2.1430022344099125e-05, + "loss": 0.105, "step": 3360 }, { - "epoch": 4.79, - "learning_rate": 2.8923660502607873e-05, - "loss": 0.0948, + "epoch": 6.16, + "learning_rate": 2.132845825716027e-05, + "loss": 0.1064, "step": 3370 }, { - "epoch": 4.81, - "learning_rate": 2.8844634107792e-05, - "loss": 0.1191, + "epoch": 6.18, + "learning_rate": 2.1226894170221413e-05, + "loss": 0.1013, "step": 3380 }, { - "epoch": 4.82, - "learning_rate": 2.8765607712976133e-05, - "loss": 0.1014, + "epoch": 6.2, + "learning_rate": 2.112533008328255e-05, + "loss": 0.1113, "step": 3390 }, { - "epoch": 4.83, - "learning_rate": 2.8686581318160265e-05, - "loss": 0.1237, + "epoch": 6.22, + "learning_rate": 2.1023765996343695e-05, + "loss": 0.1065, "step": 3400 }, { - "epoch": 4.85, - "learning_rate": 2.86075549233444e-05, - "loss": 0.1355, + "epoch": 6.23, + "learning_rate": 2.0922201909404835e-05, + "loss": 0.1014, "step": 3410 }, { - "epoch": 4.86, - "learning_rate": 2.852852852852853e-05, - "loss": 0.1327, + "epoch": 6.25, + "learning_rate": 2.0820637822465976e-05, + "loss": 0.1169, "step": 3420 }, { - "epoch": 4.88, - "learning_rate": 2.8449502133712663e-05, - "loss": 0.112, + "epoch": 6.27, + "learning_rate": 2.071907373552712e-05, + "loss": 0.1086, "step": 3430 }, { - "epoch": 4.89, - "learning_rate": 2.837047573889679e-05, - "loss": 0.1231, + "epoch": 6.29, + "learning_rate": 2.061750964858826e-05, + "loss": 0.125, "step": 3440 }, { - "epoch": 4.91, - "learning_rate": 2.8291449344080923e-05, - "loss": 0.1291, + "epoch": 6.31, + "learning_rate": 2.05159455616494e-05, + "loss": 0.089, "step": 3450 }, { - "epoch": 4.92, - "learning_rate": 2.8212422949265055e-05, - "loss": 0.1121, + "epoch": 6.33, + "learning_rate": 2.0414381474710542e-05, + "loss": 0.1056, "step": 3460 }, { - "epoch": 4.93, - "learning_rate": 2.8133396554449187e-05, - "loss": 0.1234, + "epoch": 6.34, + "learning_rate": 2.0312817387771686e-05, + "loss": 0.0981, "step": 3470 }, { - "epoch": 4.95, - "learning_rate": 2.8054370159633318e-05, - "loss": 0.0887, + "epoch": 6.36, + "learning_rate": 2.0211253300832827e-05, + "loss": 0.1178, "step": 3480 }, { - "epoch": 4.96, - "learning_rate": 2.7975343764817453e-05, - "loss": 0.1136, + "epoch": 6.38, + "learning_rate": 2.0109689213893968e-05, + "loss": 0.1065, "step": 3490 }, { - "epoch": 4.98, - "learning_rate": 2.7896317370001585e-05, - "loss": 0.1393, + "epoch": 6.4, + "learning_rate": 2.000812512695511e-05, + "loss": 0.1219, "step": 3500 }, { - "epoch": 4.99, - "learning_rate": 2.781729097518571e-05, - "loss": 0.1008, + "epoch": 6.42, + "learning_rate": 1.990656104001625e-05, + "loss": 0.1114, "step": 3510 }, { - "epoch": 5.0, - "eval_accuracy": 0.9824, - "eval_loss": 0.0443880558013916, - "eval_runtime": 59.6491, - "eval_samples_per_second": 167.647, - "eval_steps_per_second": 5.247, - "step": 3516 - }, - { - "epoch": 5.01, - "learning_rate": 2.7738264580369845e-05, - "loss": 0.1171, + "epoch": 6.44, + "learning_rate": 1.9804996953077393e-05, + "loss": 0.1034, "step": 3520 }, { - "epoch": 5.02, - "learning_rate": 2.7659238185553977e-05, - "loss": 0.1185, + "epoch": 6.45, + "learning_rate": 1.9703432866138537e-05, + "loss": 0.1159, "step": 3530 }, { - "epoch": 5.03, - "learning_rate": 2.758021179073811e-05, - "loss": 0.1209, + "epoch": 6.47, + "learning_rate": 1.9601868779199674e-05, + "loss": 0.0968, "step": 3540 }, { - "epoch": 5.05, - "learning_rate": 2.750118539592224e-05, - "loss": 0.1207, + "epoch": 6.49, + "learning_rate": 1.950030469226082e-05, + "loss": 0.108, "step": 3550 }, { - "epoch": 5.06, - "learning_rate": 2.7422159001106372e-05, - "loss": 0.1079, + "epoch": 6.51, + "learning_rate": 1.939874060532196e-05, + "loss": 0.1239, "step": 3560 }, { - "epoch": 5.08, - "learning_rate": 2.73431326062905e-05, - "loss": 0.1204, + "epoch": 6.53, + "learning_rate": 1.92971765183831e-05, + "loss": 0.0928, "step": 3570 }, { - "epoch": 5.09, - "learning_rate": 2.7264106211474632e-05, - "loss": 0.112, + "epoch": 6.54, + "learning_rate": 1.9195612431444244e-05, + "loss": 0.099, "step": 3580 }, { - "epoch": 5.1, - "learning_rate": 2.7185079816658763e-05, - "loss": 0.1139, + "epoch": 6.56, + "learning_rate": 1.9094048344505384e-05, + "loss": 0.1039, "step": 3590 }, { - "epoch": 5.12, - "learning_rate": 2.71060534218429e-05, - "loss": 0.1126, + "epoch": 6.58, + "learning_rate": 1.8992484257566525e-05, + "loss": 0.0904, "step": 3600 }, { - "epoch": 5.13, - "learning_rate": 2.702702702702703e-05, - "loss": 0.1355, + "epoch": 6.6, + "learning_rate": 1.8890920170627666e-05, + "loss": 0.1183, "step": 3610 }, { - "epoch": 5.15, - "learning_rate": 2.6948000632211162e-05, - "loss": 0.1107, + "epoch": 6.62, + "learning_rate": 1.878935608368881e-05, + "loss": 0.1198, "step": 3620 }, { - "epoch": 5.16, - "learning_rate": 2.686897423739529e-05, - "loss": 0.0918, + "epoch": 6.64, + "learning_rate": 1.868779199674995e-05, + "loss": 0.1213, "step": 3630 }, { - "epoch": 5.18, - "learning_rate": 2.6789947842579422e-05, - "loss": 0.1175, + "epoch": 6.65, + "learning_rate": 1.858622790981109e-05, + "loss": 0.0993, "step": 3640 }, { - "epoch": 5.19, - "learning_rate": 2.6710921447763554e-05, - "loss": 0.1131, + "epoch": 6.67, + "learning_rate": 1.8484663822872232e-05, + "loss": 0.1091, "step": 3650 }, { - "epoch": 5.2, - "learning_rate": 2.6631895052947685e-05, - "loss": 0.1271, + "epoch": 6.69, + "learning_rate": 1.8383099735933376e-05, + "loss": 0.1043, "step": 3660 }, { - "epoch": 5.22, - "learning_rate": 2.655286865813182e-05, - "loss": 0.1262, + "epoch": 6.71, + "learning_rate": 1.8281535648994517e-05, + "loss": 0.1065, "step": 3670 }, { - "epoch": 5.23, - "learning_rate": 2.6473842263315952e-05, - "loss": 0.1034, + "epoch": 6.73, + "learning_rate": 1.8179971562055657e-05, + "loss": 0.1055, "step": 3680 }, { - "epoch": 5.25, - "learning_rate": 2.6394815868500077e-05, - "loss": 0.1266, + "epoch": 6.75, + "learning_rate": 1.8078407475116798e-05, + "loss": 0.1151, "step": 3690 }, { - "epoch": 5.26, - "learning_rate": 2.6315789473684212e-05, - "loss": 0.1149, + "epoch": 6.76, + "learning_rate": 1.7976843388177942e-05, + "loss": 0.0938, "step": 3700 }, { - "epoch": 5.28, - "learning_rate": 2.6236763078868344e-05, - "loss": 0.1187, + "epoch": 6.78, + "learning_rate": 1.7875279301239083e-05, + "loss": 0.113, "step": 3710 }, { - "epoch": 5.29, - "learning_rate": 2.6157736684052475e-05, - "loss": 0.1143, + "epoch": 6.8, + "learning_rate": 1.7773715214300223e-05, + "loss": 0.1068, "step": 3720 }, { - "epoch": 5.3, - "learning_rate": 2.6078710289236607e-05, - "loss": 0.1066, + "epoch": 6.82, + "learning_rate": 1.7672151127361367e-05, + "loss": 0.0978, "step": 3730 }, { - "epoch": 5.32, - "learning_rate": 2.599968389442074e-05, - "loss": 0.0993, + "epoch": 6.84, + "learning_rate": 1.7570587040422508e-05, + "loss": 0.1129, "step": 3740 }, { - "epoch": 5.33, - "learning_rate": 2.5920657499604867e-05, - "loss": 0.1201, + "epoch": 6.86, + "learning_rate": 1.746902295348365e-05, + "loss": 0.1069, "step": 3750 }, { - "epoch": 5.35, - "learning_rate": 2.5841631104789e-05, - "loss": 0.0983, + "epoch": 6.87, + "learning_rate": 1.7367458866544793e-05, + "loss": 0.111, "step": 3760 }, { - "epoch": 5.36, - "learning_rate": 2.576260470997313e-05, - "loss": 0.1202, + "epoch": 6.89, + "learning_rate": 1.726589477960593e-05, + "loss": 0.0973, "step": 3770 }, { - "epoch": 5.38, - "learning_rate": 2.5683578315157265e-05, - "loss": 0.0995, + "epoch": 6.91, + "learning_rate": 1.7164330692667074e-05, + "loss": 0.1002, "step": 3780 }, { - "epoch": 5.39, - "learning_rate": 2.5604551920341397e-05, - "loss": 0.1016, + "epoch": 6.93, + "learning_rate": 1.7062766605728215e-05, + "loss": 0.1095, "step": 3790 }, { - "epoch": 5.4, - "learning_rate": 2.552552552552553e-05, - "loss": 0.1155, + "epoch": 6.95, + "learning_rate": 1.6961202518789355e-05, + "loss": 0.1023, "step": 3800 }, { - "epoch": 5.42, - "learning_rate": 2.5446499130709657e-05, - "loss": 0.129, + "epoch": 6.97, + "learning_rate": 1.68596384318505e-05, + "loss": 0.0972, "step": 3810 }, { - "epoch": 5.43, - "learning_rate": 2.536747273589379e-05, - "loss": 0.1278, + "epoch": 6.98, + "learning_rate": 1.675807434491164e-05, + "loss": 0.1047, "step": 3820 }, { - "epoch": 5.45, - "learning_rate": 2.528844634107792e-05, - "loss": 0.1111, + "epoch": 7.0, + "eval_accuracy": 0.9833333333333333, + "eval_loss": 0.04612118750810623, + "eval_runtime": 107.2386, + "eval_samples_per_second": 279.75, + "eval_steps_per_second": 8.747, + "step": 3829 + }, + { + "epoch": 7.0, + "learning_rate": 1.665651025797278e-05, + "loss": 0.1033, "step": 3830 }, { - "epoch": 5.46, - "learning_rate": 2.5209419946262052e-05, - "loss": 0.1067, + "epoch": 7.02, + "learning_rate": 1.655494617103392e-05, + "loss": 0.1108, "step": 3840 }, { - "epoch": 5.47, - "learning_rate": 2.5130393551446184e-05, - "loss": 0.1085, + "epoch": 7.04, + "learning_rate": 1.6453382084095066e-05, + "loss": 0.1146, "step": 3850 }, { - "epoch": 5.49, - "learning_rate": 2.505136715663032e-05, - "loss": 0.1171, + "epoch": 7.06, + "learning_rate": 1.6351817997156206e-05, + "loss": 0.0947, "step": 3860 }, { - "epoch": 5.5, - "learning_rate": 2.4972340761814447e-05, - "loss": 0.1067, + "epoch": 7.07, + "learning_rate": 1.6250253910217347e-05, + "loss": 0.101, "step": 3870 }, { - "epoch": 5.52, - "learning_rate": 2.489331436699858e-05, - "loss": 0.1027, + "epoch": 7.09, + "learning_rate": 1.614868982327849e-05, + "loss": 0.114, "step": 3880 }, { - "epoch": 5.53, - "learning_rate": 2.481428797218271e-05, - "loss": 0.1487, + "epoch": 7.11, + "learning_rate": 1.6047125736339628e-05, + "loss": 0.0816, "step": 3890 }, { - "epoch": 5.55, - "learning_rate": 2.4735261577366842e-05, - "loss": 0.1243, + "epoch": 7.13, + "learning_rate": 1.5945561649400772e-05, + "loss": 0.1092, "step": 3900 }, { - "epoch": 5.56, - "learning_rate": 2.4656235182550974e-05, - "loss": 0.0905, + "epoch": 7.15, + "learning_rate": 1.5843997562461916e-05, + "loss": 0.1109, "step": 3910 }, { - "epoch": 5.57, - "learning_rate": 2.4577208787735102e-05, - "loss": 0.1274, + "epoch": 7.17, + "learning_rate": 1.5742433475523054e-05, + "loss": 0.1223, "step": 3920 }, { - "epoch": 5.59, - "learning_rate": 2.4498182392919237e-05, - "loss": 0.1185, + "epoch": 7.18, + "learning_rate": 1.5640869388584198e-05, + "loss": 0.0923, "step": 3930 }, { - "epoch": 5.6, - "learning_rate": 2.441915599810337e-05, - "loss": 0.1153, + "epoch": 7.2, + "learning_rate": 1.553930530164534e-05, + "loss": 0.0975, "step": 3940 }, { - "epoch": 5.62, - "learning_rate": 2.4340129603287497e-05, - "loss": 0.1153, + "epoch": 7.22, + "learning_rate": 1.543774121470648e-05, + "loss": 0.1179, "step": 3950 }, { - "epoch": 5.63, - "learning_rate": 2.426110320847163e-05, - "loss": 0.1229, + "epoch": 7.24, + "learning_rate": 1.5336177127767623e-05, + "loss": 0.1035, "step": 3960 }, { - "epoch": 5.65, - "learning_rate": 2.4182076813655764e-05, - "loss": 0.1115, + "epoch": 7.26, + "learning_rate": 1.5234613040828762e-05, + "loss": 0.1164, "step": 3970 }, { - "epoch": 5.66, - "learning_rate": 2.4103050418839892e-05, - "loss": 0.1222, + "epoch": 7.28, + "learning_rate": 1.5133048953889906e-05, + "loss": 0.0975, "step": 3980 }, { - "epoch": 5.67, - "learning_rate": 2.4024024024024024e-05, - "loss": 0.1335, + "epoch": 7.29, + "learning_rate": 1.5031484866951045e-05, + "loss": 0.1078, "step": 3990 }, { - "epoch": 5.69, - "learning_rate": 2.3944997629208156e-05, - "loss": 0.1141, + "epoch": 7.31, + "learning_rate": 1.4929920780012188e-05, + "loss": 0.1012, "step": 4000 }, { - "epoch": 5.7, - "learning_rate": 2.3865971234392288e-05, - "loss": 0.111, + "epoch": 7.33, + "learning_rate": 1.4828356693073332e-05, + "loss": 0.0958, "step": 4010 }, { - "epoch": 5.72, - "learning_rate": 2.378694483957642e-05, - "loss": 0.1009, + "epoch": 7.35, + "learning_rate": 1.472679260613447e-05, + "loss": 0.1068, "step": 4020 }, { - "epoch": 5.73, - "learning_rate": 2.370791844476055e-05, - "loss": 0.1076, + "epoch": 7.37, + "learning_rate": 1.4625228519195613e-05, + "loss": 0.0858, "step": 4030 }, { - "epoch": 5.74, - "learning_rate": 2.3628892049944683e-05, - "loss": 0.0905, + "epoch": 7.39, + "learning_rate": 1.4523664432256754e-05, + "loss": 0.0975, "step": 4040 }, { - "epoch": 5.76, - "learning_rate": 2.3549865655128814e-05, - "loss": 0.1196, + "epoch": 7.4, + "learning_rate": 1.4422100345317896e-05, + "loss": 0.097, "step": 4050 }, { - "epoch": 5.77, - "learning_rate": 2.3470839260312946e-05, - "loss": 0.1057, + "epoch": 7.42, + "learning_rate": 1.4320536258379038e-05, + "loss": 0.0943, "step": 4060 }, { - "epoch": 5.79, - "learning_rate": 2.3391812865497074e-05, - "loss": 0.1076, + "epoch": 7.44, + "learning_rate": 1.4218972171440179e-05, + "loss": 0.0925, "step": 4070 }, { - "epoch": 5.8, - "learning_rate": 2.331278647068121e-05, - "loss": 0.1077, + "epoch": 7.46, + "learning_rate": 1.4117408084501321e-05, + "loss": 0.109, "step": 4080 }, { - "epoch": 5.82, - "learning_rate": 2.323376007586534e-05, - "loss": 0.1057, + "epoch": 7.48, + "learning_rate": 1.4015843997562464e-05, + "loss": 0.0927, "step": 4090 }, { - "epoch": 5.83, - "learning_rate": 2.315473368104947e-05, - "loss": 0.1254, + "epoch": 7.5, + "learning_rate": 1.3914279910623604e-05, + "loss": 0.1082, "step": 4100 }, { - "epoch": 5.84, - "learning_rate": 2.30757072862336e-05, - "loss": 0.1015, + "epoch": 7.51, + "learning_rate": 1.3812715823684747e-05, + "loss": 0.1114, "step": 4110 }, { - "epoch": 5.86, - "learning_rate": 2.2996680891417736e-05, - "loss": 0.1059, + "epoch": 7.53, + "learning_rate": 1.3711151736745886e-05, + "loss": 0.1156, "step": 4120 }, { - "epoch": 5.87, - "learning_rate": 2.2917654496601868e-05, - "loss": 0.1103, + "epoch": 7.55, + "learning_rate": 1.360958764980703e-05, + "loss": 0.0884, "step": 4130 }, { - "epoch": 5.89, - "learning_rate": 2.2838628101785996e-05, - "loss": 0.1141, + "epoch": 7.57, + "learning_rate": 1.3508023562868172e-05, + "loss": 0.0813, "step": 4140 }, { - "epoch": 5.9, - "learning_rate": 2.275960170697013e-05, - "loss": 0.1046, + "epoch": 7.59, + "learning_rate": 1.3406459475929311e-05, + "loss": 0.0929, "step": 4150 }, { - "epoch": 5.92, - "learning_rate": 2.2680575312154263e-05, - "loss": 0.1228, + "epoch": 7.61, + "learning_rate": 1.3304895388990453e-05, + "loss": 0.0768, "step": 4160 }, { - "epoch": 5.93, - "learning_rate": 2.260154891733839e-05, - "loss": 0.099, + "epoch": 7.62, + "learning_rate": 1.3203331302051594e-05, + "loss": 0.0982, "step": 4170 }, { - "epoch": 5.94, - "learning_rate": 2.2522522522522523e-05, - "loss": 0.1156, + "epoch": 7.64, + "learning_rate": 1.3101767215112737e-05, + "loss": 0.0945, "step": 4180 }, { - "epoch": 5.96, - "learning_rate": 2.2443496127706658e-05, - "loss": 0.0834, + "epoch": 7.66, + "learning_rate": 1.3000203128173879e-05, + "loss": 0.1056, "step": 4190 }, { - "epoch": 5.97, - "learning_rate": 2.2364469732890786e-05, - "loss": 0.0929, + "epoch": 7.68, + "learning_rate": 1.289863904123502e-05, + "loss": 0.1006, "step": 4200 }, { - "epoch": 5.99, - "learning_rate": 2.2285443338074918e-05, - "loss": 0.1189, + "epoch": 7.7, + "learning_rate": 1.2797074954296162e-05, + "loss": 0.0994, "step": 4210 }, { - "epoch": 6.0, - "eval_accuracy": 0.977, - "eval_loss": 0.06354519724845886, - "eval_runtime": 59.4966, - "eval_samples_per_second": 168.077, - "eval_steps_per_second": 5.261, - "step": 4219 + "epoch": 7.71, + "learning_rate": 1.2695510867357303e-05, + "loss": 0.1097, + "step": 4220 }, { - "epoch": 6.0, - "learning_rate": 2.220641694325905e-05, - "loss": 0.0914, - "step": 4220 - }, - { - "epoch": 6.01, - "learning_rate": 2.212739054844318e-05, - "loss": 0.113, + "epoch": 7.73, + "learning_rate": 1.2593946780418445e-05, + "loss": 0.0899, "step": 4230 }, { - "epoch": 6.03, - "learning_rate": 2.2048364153627313e-05, - "loss": 0.1103, + "epoch": 7.75, + "learning_rate": 1.2492382693479586e-05, + "loss": 0.0872, "step": 4240 }, { - "epoch": 6.04, - "learning_rate": 2.1969337758811445e-05, - "loss": 0.1226, + "epoch": 7.77, + "learning_rate": 1.2390818606540728e-05, + "loss": 0.1199, "step": 4250 }, { - "epoch": 6.06, - "learning_rate": 2.1890311363995576e-05, - "loss": 0.1094, + "epoch": 7.79, + "learning_rate": 1.228925451960187e-05, + "loss": 0.1142, "step": 4260 }, { - "epoch": 6.07, - "learning_rate": 2.1811284969179708e-05, - "loss": 0.119, + "epoch": 7.81, + "learning_rate": 1.2187690432663011e-05, + "loss": 0.1281, "step": 4270 }, { - "epoch": 6.09, - "learning_rate": 2.173225857436384e-05, - "loss": 0.113, + "epoch": 7.82, + "learning_rate": 1.2086126345724152e-05, + "loss": 0.1163, "step": 4280 }, { - "epoch": 6.1, - "learning_rate": 2.1653232179547968e-05, - "loss": 0.1079, + "epoch": 7.84, + "learning_rate": 1.1984562258785294e-05, + "loss": 0.0965, "step": 4290 }, { - "epoch": 6.11, - "learning_rate": 2.1574205784732103e-05, - "loss": 0.1015, + "epoch": 7.86, + "learning_rate": 1.1882998171846435e-05, + "loss": 0.1009, "step": 4300 }, { - "epoch": 6.13, - "learning_rate": 2.1495179389916235e-05, - "loss": 0.108, + "epoch": 7.88, + "learning_rate": 1.1781434084907577e-05, + "loss": 0.0951, "step": 4310 }, { - "epoch": 6.14, - "learning_rate": 2.1416152995100363e-05, - "loss": 0.0954, + "epoch": 7.9, + "learning_rate": 1.167986999796872e-05, + "loss": 0.093, "step": 4320 }, { - "epoch": 6.16, - "learning_rate": 2.1337126600284495e-05, - "loss": 0.093, + "epoch": 7.92, + "learning_rate": 1.157830591102986e-05, + "loss": 0.0951, "step": 4330 }, { - "epoch": 6.17, - "learning_rate": 2.125810020546863e-05, - "loss": 0.0899, + "epoch": 7.93, + "learning_rate": 1.1476741824091e-05, + "loss": 0.0957, "step": 4340 }, { - "epoch": 6.19, - "learning_rate": 2.1179073810652758e-05, - "loss": 0.1142, + "epoch": 7.95, + "learning_rate": 1.1375177737152143e-05, + "loss": 0.09, "step": 4350 }, { - "epoch": 6.2, - "learning_rate": 2.110004741583689e-05, - "loss": 0.1046, + "epoch": 7.97, + "learning_rate": 1.1273613650213286e-05, + "loss": 0.1188, "step": 4360 }, { - "epoch": 6.21, - "learning_rate": 2.102102102102102e-05, - "loss": 0.1051, + "epoch": 7.99, + "learning_rate": 1.1172049563274426e-05, + "loss": 0.1079, "step": 4370 }, { - "epoch": 6.23, - "learning_rate": 2.0941994626205153e-05, - "loss": 0.1096, + "epoch": 8.0, + "eval_accuracy": 0.9829, + "eval_loss": 0.045376840978860855, + "eval_runtime": 107.845, + "eval_samples_per_second": 278.177, + "eval_steps_per_second": 8.698, + "step": 4376 + }, + { + "epoch": 8.01, + "learning_rate": 1.1070485476335569e-05, + "loss": 0.0779, "step": 4380 }, { - "epoch": 6.24, - "learning_rate": 2.0862968231389285e-05, - "loss": 0.1282, + "epoch": 8.03, + "learning_rate": 1.096892138939671e-05, + "loss": 0.1101, "step": 4390 }, { - "epoch": 6.26, - "learning_rate": 2.0783941836573417e-05, - "loss": 0.1124, + "epoch": 8.04, + "learning_rate": 1.0867357302457852e-05, + "loss": 0.0848, "step": 4400 }, { - "epoch": 6.27, - "learning_rate": 2.070491544175755e-05, - "loss": 0.0819, + "epoch": 8.06, + "learning_rate": 1.0765793215518994e-05, + "loss": 0.0859, "step": 4410 }, { - "epoch": 6.29, - "learning_rate": 2.062588904694168e-05, - "loss": 0.0963, + "epoch": 8.08, + "learning_rate": 1.0664229128580135e-05, + "loss": 0.0805, "step": 4420 }, { - "epoch": 6.3, - "learning_rate": 2.0546862652125812e-05, - "loss": 0.1037, + "epoch": 8.1, + "learning_rate": 1.0562665041641275e-05, + "loss": 0.0838, "step": 4430 }, { - "epoch": 6.31, - "learning_rate": 2.046783625730994e-05, - "loss": 0.0856, + "epoch": 8.12, + "learning_rate": 1.0461100954702418e-05, + "loss": 0.1039, "step": 4440 }, { - "epoch": 6.33, - "learning_rate": 2.0388809862494075e-05, - "loss": 0.1141, + "epoch": 8.14, + "learning_rate": 1.035953686776356e-05, + "loss": 0.0895, "step": 4450 }, { - "epoch": 6.34, - "learning_rate": 2.0309783467678207e-05, - "loss": 0.112, + "epoch": 8.15, + "learning_rate": 1.02579727808247e-05, + "loss": 0.095, "step": 4460 }, { - "epoch": 6.36, - "learning_rate": 2.0230757072862335e-05, - "loss": 0.1134, + "epoch": 8.17, + "learning_rate": 1.0156408693885843e-05, + "loss": 0.0947, "step": 4470 }, { - "epoch": 6.37, - "learning_rate": 2.0151730678046467e-05, - "loss": 0.0848, + "epoch": 8.19, + "learning_rate": 1.0054844606946984e-05, + "loss": 0.0817, "step": 4480 }, { - "epoch": 6.38, - "learning_rate": 2.0072704283230602e-05, - "loss": 0.1022, + "epoch": 8.21, + "learning_rate": 9.953280520008124e-06, + "loss": 0.102, "step": 4490 }, { - "epoch": 6.4, - "learning_rate": 1.999367788841473e-05, - "loss": 0.091, + "epoch": 8.23, + "learning_rate": 9.851716433069269e-06, + "loss": 0.0909, "step": 4500 }, { - "epoch": 6.41, - "learning_rate": 1.9914651493598862e-05, - "loss": 0.0989, + "epoch": 8.24, + "learning_rate": 9.75015234613041e-06, + "loss": 0.0908, "step": 4510 }, { - "epoch": 6.43, - "learning_rate": 1.9835625098782994e-05, - "loss": 0.1159, + "epoch": 8.26, + "learning_rate": 9.64858825919155e-06, + "loss": 0.1009, "step": 4520 }, { - "epoch": 6.44, - "learning_rate": 1.9756598703967125e-05, - "loss": 0.1183, + "epoch": 8.28, + "learning_rate": 9.547024172252692e-06, + "loss": 0.0984, "step": 4530 }, { - "epoch": 6.46, - "learning_rate": 1.9677572309151257e-05, - "loss": 0.0916, + "epoch": 8.3, + "learning_rate": 9.445460085313833e-06, + "loss": 0.0914, "step": 4540 }, { - "epoch": 6.47, - "learning_rate": 1.959854591433539e-05, - "loss": 0.1234, + "epoch": 8.32, + "learning_rate": 9.343895998374975e-06, + "loss": 0.0986, "step": 4550 }, { - "epoch": 6.48, - "learning_rate": 1.951951951951952e-05, - "loss": 0.1049, + "epoch": 8.34, + "learning_rate": 9.242331911436116e-06, + "loss": 0.0975, "step": 4560 }, { - "epoch": 6.5, - "learning_rate": 1.9440493124703652e-05, - "loss": 0.1282, + "epoch": 8.35, + "learning_rate": 9.140767824497258e-06, + "loss": 0.1058, "step": 4570 }, { - "epoch": 6.51, - "learning_rate": 1.9361466729887784e-05, - "loss": 0.0907, + "epoch": 8.37, + "learning_rate": 9.039203737558399e-06, + "loss": 0.0891, "step": 4580 }, { - "epoch": 6.53, - "learning_rate": 1.9282440335071915e-05, - "loss": 0.0939, + "epoch": 8.39, + "learning_rate": 8.937639650619541e-06, + "loss": 0.0834, "step": 4590 }, { - "epoch": 6.54, - "learning_rate": 1.9203413940256047e-05, - "loss": 0.1022, + "epoch": 8.41, + "learning_rate": 8.836075563680684e-06, + "loss": 0.0894, "step": 4600 }, { - "epoch": 6.56, - "learning_rate": 1.912438754544018e-05, - "loss": 0.1305, + "epoch": 8.43, + "learning_rate": 8.734511476741824e-06, + "loss": 0.102, "step": 4610 }, { - "epoch": 6.57, - "learning_rate": 1.904536115062431e-05, - "loss": 0.0918, + "epoch": 8.45, + "learning_rate": 8.632947389802965e-06, + "loss": 0.0703, "step": 4620 }, { - "epoch": 6.58, - "learning_rate": 1.896633475580844e-05, - "loss": 0.0948, + "epoch": 8.46, + "learning_rate": 8.531383302864107e-06, + "loss": 0.0821, "step": 4630 }, { - "epoch": 6.6, - "learning_rate": 1.8887308360992574e-05, - "loss": 0.0913, + "epoch": 8.48, + "learning_rate": 8.42981921592525e-06, + "loss": 0.093, "step": 4640 }, { - "epoch": 6.61, - "learning_rate": 1.8808281966176705e-05, - "loss": 0.0961, + "epoch": 8.5, + "learning_rate": 8.32825512898639e-06, + "loss": 0.0901, "step": 4650 }, { - "epoch": 6.63, - "learning_rate": 1.8729255571360834e-05, - "loss": 0.119, + "epoch": 8.52, + "learning_rate": 8.226691042047533e-06, + "loss": 0.0958, "step": 4660 }, { - "epoch": 6.64, - "learning_rate": 1.8650229176544965e-05, - "loss": 0.119, + "epoch": 8.54, + "learning_rate": 8.125126955108673e-06, + "loss": 0.0978, "step": 4670 }, { - "epoch": 6.65, - "learning_rate": 1.85712027817291e-05, - "loss": 0.1092, + "epoch": 8.56, + "learning_rate": 8.023562868169814e-06, + "loss": 0.1037, "step": 4680 }, { - "epoch": 6.67, - "learning_rate": 1.849217638691323e-05, - "loss": 0.1111, + "epoch": 8.57, + "learning_rate": 7.921998781230958e-06, + "loss": 0.0704, "step": 4690 }, { - "epoch": 6.68, - "learning_rate": 1.841314999209736e-05, - "loss": 0.0882, + "epoch": 8.59, + "learning_rate": 7.820434694292099e-06, + "loss": 0.097, "step": 4700 }, { - "epoch": 6.7, - "learning_rate": 1.8334123597281496e-05, - "loss": 0.0968, + "epoch": 8.61, + "learning_rate": 7.71887060735324e-06, + "loss": 0.0957, "step": 4710 }, { - "epoch": 6.71, - "learning_rate": 1.8255097202465624e-05, - "loss": 0.1249, + "epoch": 8.63, + "learning_rate": 7.617306520414381e-06, + "loss": 0.0784, "step": 4720 }, { - "epoch": 6.73, - "learning_rate": 1.8176070807649756e-05, - "loss": 0.125, + "epoch": 8.65, + "learning_rate": 7.5157424334755226e-06, + "loss": 0.0959, "step": 4730 }, { - "epoch": 6.74, - "learning_rate": 1.8097044412833887e-05, - "loss": 0.0981, + "epoch": 8.67, + "learning_rate": 7.414178346536666e-06, + "loss": 0.1128, "step": 4740 }, { - "epoch": 6.75, - "learning_rate": 1.801801801801802e-05, - "loss": 0.1078, + "epoch": 8.68, + "learning_rate": 7.3126142595978065e-06, + "loss": 0.0889, "step": 4750 }, { - "epoch": 6.77, - "learning_rate": 1.793899162320215e-05, - "loss": 0.0949, + "epoch": 8.7, + "learning_rate": 7.211050172658948e-06, + "loss": 0.0939, "step": 4760 }, { - "epoch": 6.78, - "learning_rate": 1.7859965228386282e-05, - "loss": 0.0952, + "epoch": 8.72, + "learning_rate": 7.1094860857200895e-06, + "loss": 0.0994, "step": 4770 }, { - "epoch": 6.8, - "learning_rate": 1.7780938833570414e-05, - "loss": 0.0863, + "epoch": 8.74, + "learning_rate": 7.007921998781232e-06, + "loss": 0.1026, "step": 4780 }, { - "epoch": 6.81, - "learning_rate": 1.7701912438754546e-05, - "loss": 0.0878, + "epoch": 8.76, + "learning_rate": 6.906357911842373e-06, + "loss": 0.0855, "step": 4790 }, { - "epoch": 6.83, - "learning_rate": 1.7622886043938677e-05, - "loss": 0.0944, + "epoch": 8.78, + "learning_rate": 6.804793824903515e-06, + "loss": 0.097, "step": 4800 }, { - "epoch": 6.84, - "learning_rate": 1.7543859649122806e-05, - "loss": 0.0911, + "epoch": 8.79, + "learning_rate": 6.7032297379646556e-06, + "loss": 0.0877, "step": 4810 }, { - "epoch": 6.85, - "learning_rate": 1.746483325430694e-05, - "loss": 0.0844, + "epoch": 8.81, + "learning_rate": 6.601665651025797e-06, + "loss": 0.1014, "step": 4820 }, { - "epoch": 6.87, - "learning_rate": 1.7385806859491072e-05, - "loss": 0.1032, + "epoch": 8.83, + "learning_rate": 6.5001015640869394e-06, + "loss": 0.0919, "step": 4830 }, { - "epoch": 6.88, - "learning_rate": 1.73067804646752e-05, - "loss": 0.1096, + "epoch": 8.85, + "learning_rate": 6.398537477148081e-06, + "loss": 0.1037, "step": 4840 }, { - "epoch": 6.9, - "learning_rate": 1.7227754069859332e-05, - "loss": 0.0933, + "epoch": 8.87, + "learning_rate": 6.2969733902092225e-06, + "loss": 0.0964, "step": 4850 }, { - "epoch": 6.91, - "learning_rate": 1.7148727675043468e-05, - "loss": 0.1098, + "epoch": 8.88, + "learning_rate": 6.195409303270364e-06, + "loss": 0.0892, "step": 4860 }, { - "epoch": 6.92, - "learning_rate": 1.7069701280227596e-05, - "loss": 0.0994, + "epoch": 8.9, + "learning_rate": 6.0938452163315055e-06, + "loss": 0.1053, "step": 4870 }, { - "epoch": 6.94, - "learning_rate": 1.6990674885411728e-05, - "loss": 0.0983, + "epoch": 8.92, + "learning_rate": 5.992281129392647e-06, + "loss": 0.0824, "step": 4880 }, { - "epoch": 6.95, - "learning_rate": 1.691164849059586e-05, - "loss": 0.1076, + "epoch": 8.94, + "learning_rate": 5.8907170424537886e-06, + "loss": 0.0837, "step": 4890 }, { - "epoch": 6.97, - "learning_rate": 1.683262209577999e-05, - "loss": 0.0996, + "epoch": 8.96, + "learning_rate": 5.78915295551493e-06, + "loss": 0.0976, "step": 4900 }, { - "epoch": 6.98, - "learning_rate": 1.6753595700964123e-05, - "loss": 0.0863, + "epoch": 8.98, + "learning_rate": 5.687588868576072e-06, + "loss": 0.0895, "step": 4910 }, { - "epoch": 7.0, - "learning_rate": 1.6674569306148254e-05, - "loss": 0.0936, + "epoch": 8.99, + "learning_rate": 5.586024781637213e-06, + "loss": 0.0805, "step": 4920 }, { - "epoch": 7.0, - "eval_accuracy": 0.9789, - "eval_loss": 0.0534820482134819, - "eval_runtime": 59.7551, - "eval_samples_per_second": 167.35, - "eval_steps_per_second": 5.238, - "step": 4922 + "epoch": 9.0, + "eval_accuracy": 0.9790333333333333, + "eval_loss": 0.05767909437417984, + "eval_runtime": 107.4183, + "eval_samples_per_second": 279.282, + "eval_steps_per_second": 8.732, + "step": 4923 }, { - "epoch": 7.01, - "learning_rate": 1.6595542911332386e-05, - "loss": 0.1044, + "epoch": 9.01, + "learning_rate": 5.484460694698355e-06, + "loss": 0.084, "step": 4930 }, { - "epoch": 7.02, - "learning_rate": 1.6516516516516518e-05, - "loss": 0.0981, + "epoch": 9.03, + "learning_rate": 5.382896607759497e-06, + "loss": 0.083, "step": 4940 }, { - "epoch": 7.04, - "learning_rate": 1.643749012170065e-05, - "loss": 0.0757, + "epoch": 9.05, + "learning_rate": 5.281332520820638e-06, + "loss": 0.0871, "step": 4950 }, { - "epoch": 7.05, - "learning_rate": 1.6358463726884778e-05, - "loss": 0.0857, + "epoch": 9.07, + "learning_rate": 5.17976843388178e-06, + "loss": 0.0793, "step": 4960 }, { - "epoch": 7.07, - "learning_rate": 1.6279437332068913e-05, - "loss": 0.1061, + "epoch": 9.09, + "learning_rate": 5.0782043469429216e-06, + "loss": 0.0813, "step": 4970 }, { - "epoch": 7.08, - "learning_rate": 1.6200410937253044e-05, - "loss": 0.0842, + "epoch": 9.1, + "learning_rate": 4.976640260004062e-06, + "loss": 0.0886, "step": 4980 }, { - "epoch": 7.1, - "learning_rate": 1.6121384542437173e-05, - "loss": 0.0854, + "epoch": 9.12, + "learning_rate": 4.875076173065205e-06, + "loss": 0.1115, "step": 4990 }, { - "epoch": 7.11, - "learning_rate": 1.6042358147621304e-05, - "loss": 0.0872, + "epoch": 9.14, + "learning_rate": 4.773512086126346e-06, + "loss": 0.0812, "step": 5000 }, { - "epoch": 7.12, - "learning_rate": 1.596333175280544e-05, - "loss": 0.1066, + "epoch": 9.16, + "learning_rate": 4.671947999187488e-06, + "loss": 0.0768, "step": 5010 }, { - "epoch": 7.14, - "learning_rate": 1.588430535798957e-05, - "loss": 0.0951, + "epoch": 9.18, + "learning_rate": 4.570383912248629e-06, + "loss": 0.0835, "step": 5020 }, { - "epoch": 7.15, - "learning_rate": 1.58052789631737e-05, - "loss": 0.0889, + "epoch": 9.2, + "learning_rate": 4.468819825309771e-06, + "loss": 0.0817, "step": 5030 }, { - "epoch": 7.17, - "learning_rate": 1.572625256835783e-05, - "loss": 0.0975, + "epoch": 9.21, + "learning_rate": 4.367255738370912e-06, + "loss": 0.0866, "step": 5040 }, { - "epoch": 7.18, - "learning_rate": 1.5647226173541966e-05, - "loss": 0.0858, + "epoch": 9.23, + "learning_rate": 4.265691651432054e-06, + "loss": 0.0742, "step": 5050 }, { - "epoch": 7.2, - "learning_rate": 1.5568199778726095e-05, - "loss": 0.0917, + "epoch": 9.25, + "learning_rate": 4.164127564493195e-06, + "loss": 0.0755, "step": 5060 }, { - "epoch": 7.21, - "learning_rate": 1.5489173383910226e-05, - "loss": 0.079, + "epoch": 9.27, + "learning_rate": 4.062563477554337e-06, + "loss": 0.0804, "step": 5070 }, { - "epoch": 7.22, - "learning_rate": 1.5410146989094358e-05, - "loss": 0.0853, + "epoch": 9.29, + "learning_rate": 3.960999390615479e-06, + "loss": 0.0847, "step": 5080 }, { - "epoch": 7.24, - "learning_rate": 1.533112059427849e-05, - "loss": 0.1045, + "epoch": 9.31, + "learning_rate": 3.85943530367662e-06, + "loss": 0.0821, "step": 5090 }, { - "epoch": 7.25, - "learning_rate": 1.5252094199462621e-05, - "loss": 0.0895, + "epoch": 9.32, + "learning_rate": 3.7578712167377613e-06, + "loss": 0.0707, "step": 5100 }, { - "epoch": 7.27, - "learning_rate": 1.5173067804646755e-05, - "loss": 0.0892, + "epoch": 9.34, + "learning_rate": 3.6563071297989032e-06, + "loss": 0.0957, "step": 5110 }, { - "epoch": 7.28, - "learning_rate": 1.5094041409830883e-05, - "loss": 0.1013, + "epoch": 9.36, + "learning_rate": 3.5547430428600447e-06, + "loss": 0.0795, "step": 5120 }, { - "epoch": 7.29, - "learning_rate": 1.5015015015015016e-05, - "loss": 0.0946, + "epoch": 9.38, + "learning_rate": 3.4531789559211867e-06, + "loss": 0.0887, "step": 5130 }, { - "epoch": 7.31, - "learning_rate": 1.4935988620199148e-05, - "loss": 0.0878, + "epoch": 9.4, + "learning_rate": 3.3516148689823278e-06, + "loss": 0.0752, "step": 5140 }, { - "epoch": 7.32, - "learning_rate": 1.4856962225383278e-05, - "loss": 0.1028, + "epoch": 9.41, + "learning_rate": 3.2500507820434697e-06, + "loss": 0.0869, "step": 5150 }, { - "epoch": 7.34, - "learning_rate": 1.477793583056741e-05, - "loss": 0.0851, + "epoch": 9.43, + "learning_rate": 3.1484866951046112e-06, + "loss": 0.0769, "step": 5160 }, { - "epoch": 7.35, - "learning_rate": 1.4698909435751543e-05, - "loss": 0.0942, + "epoch": 9.45, + "learning_rate": 3.0469226081657528e-06, + "loss": 0.0965, "step": 5170 }, { - "epoch": 7.37, - "learning_rate": 1.4619883040935673e-05, - "loss": 0.0883, + "epoch": 9.47, + "learning_rate": 2.9453585212268943e-06, + "loss": 0.0832, "step": 5180 }, { - "epoch": 7.38, - "learning_rate": 1.4540856646119805e-05, - "loss": 0.079, + "epoch": 9.49, + "learning_rate": 2.843794434288036e-06, + "loss": 0.1018, "step": 5190 }, { - "epoch": 7.39, - "learning_rate": 1.4461830251303937e-05, - "loss": 0.0804, + "epoch": 9.51, + "learning_rate": 2.7422303473491773e-06, + "loss": 0.0818, "step": 5200 }, { - "epoch": 7.41, - "learning_rate": 1.4382803856488066e-05, - "loss": 0.0976, + "epoch": 9.52, + "learning_rate": 2.640666260410319e-06, + "loss": 0.084, "step": 5210 }, { - "epoch": 7.42, - "learning_rate": 1.43037774616722e-05, - "loss": 0.1087, + "epoch": 9.54, + "learning_rate": 2.5391021734714608e-06, + "loss": 0.0711, "step": 5220 }, { - "epoch": 7.44, - "learning_rate": 1.4224751066856332e-05, - "loss": 0.0889, + "epoch": 9.56, + "learning_rate": 2.4375380865326023e-06, + "loss": 0.0786, "step": 5230 }, { - "epoch": 7.45, - "learning_rate": 1.4145724672040462e-05, - "loss": 0.0915, + "epoch": 9.58, + "learning_rate": 2.335973999593744e-06, + "loss": 0.0875, "step": 5240 }, { - "epoch": 7.47, - "learning_rate": 1.4066698277224593e-05, - "loss": 0.0933, + "epoch": 9.6, + "learning_rate": 2.2344099126548853e-06, + "loss": 0.0948, "step": 5250 }, { - "epoch": 7.48, - "learning_rate": 1.3987671882408727e-05, - "loss": 0.1267, + "epoch": 9.62, + "learning_rate": 2.132845825716027e-06, + "loss": 0.0863, "step": 5260 }, { - "epoch": 7.49, - "learning_rate": 1.3908645487592855e-05, - "loss": 0.0947, + "epoch": 9.63, + "learning_rate": 2.0312817387771684e-06, + "loss": 0.0563, "step": 5270 }, { - "epoch": 7.51, - "learning_rate": 1.3829619092776988e-05, - "loss": 0.0952, + "epoch": 9.65, + "learning_rate": 1.92971765183831e-06, + "loss": 0.0836, "step": 5280 }, { - "epoch": 7.52, - "learning_rate": 1.375059269796112e-05, - "loss": 0.0761, + "epoch": 9.67, + "learning_rate": 1.8281535648994516e-06, + "loss": 0.0869, "step": 5290 }, { - "epoch": 7.54, - "learning_rate": 1.367156630314525e-05, - "loss": 0.0965, + "epoch": 9.69, + "learning_rate": 1.7265894779605933e-06, + "loss": 0.0961, "step": 5300 }, { - "epoch": 7.55, - "learning_rate": 1.3592539908329382e-05, - "loss": 0.1109, + "epoch": 9.71, + "learning_rate": 1.6250253910217349e-06, + "loss": 0.0811, "step": 5310 }, { - "epoch": 7.56, - "learning_rate": 1.3513513513513515e-05, - "loss": 0.0994, + "epoch": 9.73, + "learning_rate": 1.5234613040828764e-06, + "loss": 0.092, "step": 5320 }, { - "epoch": 7.58, - "learning_rate": 1.3434487118697645e-05, - "loss": 0.0951, + "epoch": 9.74, + "learning_rate": 1.421897217144018e-06, + "loss": 0.059, "step": 5330 }, { - "epoch": 7.59, - "learning_rate": 1.3355460723881777e-05, - "loss": 0.1052, + "epoch": 9.76, + "learning_rate": 1.3203331302051594e-06, + "loss": 0.0866, "step": 5340 }, { - "epoch": 7.61, - "learning_rate": 1.327643432906591e-05, - "loss": 0.0902, + "epoch": 9.78, + "learning_rate": 1.2187690432663011e-06, + "loss": 0.075, "step": 5350 }, { - "epoch": 7.62, - "learning_rate": 1.3197407934250038e-05, - "loss": 0.0858, + "epoch": 9.8, + "learning_rate": 1.1172049563274427e-06, + "loss": 0.0818, "step": 5360 }, { - "epoch": 7.64, - "learning_rate": 1.3118381539434172e-05, - "loss": 0.0908, + "epoch": 9.82, + "learning_rate": 1.0156408693885842e-06, + "loss": 0.086, "step": 5370 }, { - "epoch": 7.65, - "learning_rate": 1.3039355144618304e-05, - "loss": 0.0952, + "epoch": 9.84, + "learning_rate": 9.140767824497258e-07, + "loss": 0.0767, "step": 5380 }, { - "epoch": 7.66, - "learning_rate": 1.2960328749802434e-05, - "loss": 0.099, + "epoch": 9.85, + "learning_rate": 8.125126955108674e-07, + "loss": 0.0817, "step": 5390 }, { - "epoch": 7.68, - "learning_rate": 1.2881302354986565e-05, - "loss": 0.0982, + "epoch": 9.87, + "learning_rate": 7.10948608572009e-07, + "loss": 0.0831, "step": 5400 }, { - "epoch": 7.69, - "learning_rate": 1.2802275960170699e-05, - "loss": 0.0886, + "epoch": 9.89, + "learning_rate": 6.093845216331506e-07, + "loss": 0.0717, "step": 5410 }, { - "epoch": 7.71, - "learning_rate": 1.2723249565354829e-05, - "loss": 0.0873, + "epoch": 9.91, + "learning_rate": 5.078204346942921e-07, + "loss": 0.097, "step": 5420 }, { - "epoch": 7.72, - "learning_rate": 1.264422317053896e-05, - "loss": 0.1065, + "epoch": 9.93, + "learning_rate": 4.062563477554337e-07, + "loss": 0.0781, "step": 5430 }, { - "epoch": 7.74, - "learning_rate": 1.2565196775723092e-05, - "loss": 0.0807, + "epoch": 9.95, + "learning_rate": 3.046922608165753e-07, + "loss": 0.0792, "step": 5440 }, { - "epoch": 7.75, - "learning_rate": 1.2486170380907224e-05, - "loss": 0.0742, - "step": 5450 - }, - { - "epoch": 7.76, - "learning_rate": 1.2407143986091355e-05, - "loss": 0.0856, - "step": 5460 - }, - { - "epoch": 7.78, - "learning_rate": 1.2328117591275487e-05, - "loss": 0.1036, - "step": 5470 - }, - { - "epoch": 7.79, - "learning_rate": 1.2249091196459619e-05, - "loss": 0.0879, - "step": 5480 - }, - { - "epoch": 7.81, - "learning_rate": 1.2170064801643749e-05, - "loss": 0.1094, - "step": 5490 - }, - { - "epoch": 7.82, - "learning_rate": 1.2091038406827882e-05, - "loss": 0.0821, - "step": 5500 - }, - { - "epoch": 7.84, - "learning_rate": 1.2012012012012012e-05, - "loss": 0.0852, - "step": 5510 - }, - { - "epoch": 7.85, - "learning_rate": 1.1932985617196144e-05, - "loss": 0.105, - "step": 5520 - }, - { - "epoch": 7.86, - "learning_rate": 1.1853959222380275e-05, - "loss": 0.1152, - "step": 5530 - }, - { - "epoch": 7.88, - "learning_rate": 1.1774932827564407e-05, - "loss": 0.0926, - "step": 5540 - }, - { - "epoch": 7.89, - "learning_rate": 1.1695906432748537e-05, - "loss": 0.0987, - "step": 5550 - }, - { - "epoch": 7.91, - "learning_rate": 1.161688003793267e-05, - "loss": 0.0982, - "step": 5560 - }, - { - "epoch": 7.92, - "learning_rate": 1.15378536431168e-05, - "loss": 0.0934, - "step": 5570 - }, - { - "epoch": 7.93, - "learning_rate": 1.1458827248300934e-05, - "loss": 0.0865, - "step": 5580 - }, - { - "epoch": 7.95, - "learning_rate": 1.1379800853485066e-05, - "loss": 0.1016, - "step": 5590 - }, - { - "epoch": 7.96, - "learning_rate": 1.1300774458669196e-05, - "loss": 0.0936, - "step": 5600 - }, - { - "epoch": 7.98, - "learning_rate": 1.1221748063853329e-05, - "loss": 0.0917, - "step": 5610 - }, - { - "epoch": 7.99, - "learning_rate": 1.1142721669037459e-05, - "loss": 0.1201, - "step": 5620 - }, - { - "epoch": 8.0, - "eval_accuracy": 0.9854, - "eval_loss": 0.03756237030029297, - "eval_runtime": 59.7644, - "eval_samples_per_second": 167.324, - "eval_steps_per_second": 5.237, - "step": 5626 - }, - { - "epoch": 8.01, - "learning_rate": 1.106369527422159e-05, - "loss": 0.0872, - "step": 5630 - }, - { - "epoch": 8.02, - "learning_rate": 1.0984668879405722e-05, - "loss": 0.0914, - "step": 5640 - }, - { - "epoch": 8.03, - "learning_rate": 1.0905642484589854e-05, - "loss": 0.0965, - "step": 5650 - }, - { - "epoch": 8.05, - "learning_rate": 1.0826616089773984e-05, - "loss": 0.0961, - "step": 5660 - }, - { - "epoch": 8.06, - "learning_rate": 1.0747589694958117e-05, - "loss": 0.0823, - "step": 5670 - }, - { - "epoch": 8.08, - "learning_rate": 1.0668563300142247e-05, - "loss": 0.093, - "step": 5680 - }, - { - "epoch": 8.09, - "learning_rate": 1.0589536905326379e-05, - "loss": 0.1008, - "step": 5690 - }, - { - "epoch": 8.11, - "learning_rate": 1.051051051051051e-05, - "loss": 0.0933, - "step": 5700 - }, - { - "epoch": 8.12, - "learning_rate": 1.0431484115694642e-05, - "loss": 0.0933, - "step": 5710 - }, - { - "epoch": 8.13, - "learning_rate": 1.0352457720878774e-05, - "loss": 0.0759, - "step": 5720 - }, - { - "epoch": 8.15, - "learning_rate": 1.0273431326062906e-05, - "loss": 0.0911, - "step": 5730 - }, - { - "epoch": 8.16, - "learning_rate": 1.0194404931247038e-05, - "loss": 0.0893, - "step": 5740 - }, - { - "epoch": 8.18, - "learning_rate": 1.0115378536431168e-05, - "loss": 0.0979, - "step": 5750 - }, - { - "epoch": 8.19, - "learning_rate": 1.0036352141615301e-05, - "loss": 0.0835, - "step": 5760 - }, - { - "epoch": 8.2, - "learning_rate": 9.957325746799431e-06, - "loss": 0.08, - "step": 5770 - }, - { - "epoch": 8.22, - "learning_rate": 9.878299351983563e-06, - "loss": 0.0963, - "step": 5780 - }, - { - "epoch": 8.23, - "learning_rate": 9.799272957167694e-06, - "loss": 0.07, - "step": 5790 - }, - { - "epoch": 8.25, - "learning_rate": 9.720246562351826e-06, - "loss": 0.098, - "step": 5800 - }, - { - "epoch": 8.26, - "learning_rate": 9.641220167535958e-06, - "loss": 0.0849, - "step": 5810 - }, - { - "epoch": 8.28, - "learning_rate": 9.56219377272009e-06, - "loss": 0.0709, - "step": 5820 - }, - { - "epoch": 8.29, - "learning_rate": 9.48316737790422e-06, - "loss": 0.0807, - "step": 5830 - }, - { - "epoch": 8.3, - "learning_rate": 9.404140983088353e-06, - "loss": 0.0914, - "step": 5840 - }, - { - "epoch": 8.32, - "learning_rate": 9.325114588272483e-06, - "loss": 0.0841, - "step": 5850 - }, - { - "epoch": 8.33, - "learning_rate": 9.246088193456614e-06, - "loss": 0.072, - "step": 5860 - }, - { - "epoch": 8.35, - "learning_rate": 9.167061798640748e-06, - "loss": 0.0941, - "step": 5870 - }, - { - "epoch": 8.36, - "learning_rate": 9.088035403824878e-06, - "loss": 0.0721, - "step": 5880 - }, - { - "epoch": 8.38, - "learning_rate": 9.00900900900901e-06, - "loss": 0.0897, - "step": 5890 - }, - { - "epoch": 8.39, - "learning_rate": 8.929982614193141e-06, - "loss": 0.0899, - "step": 5900 - }, - { - "epoch": 8.4, - "learning_rate": 8.850956219377273e-06, - "loss": 0.0901, - "step": 5910 - }, - { - "epoch": 8.42, - "learning_rate": 8.771929824561403e-06, - "loss": 0.0844, - "step": 5920 - }, - { - "epoch": 8.43, - "learning_rate": 8.692903429745536e-06, - "loss": 0.0801, - "step": 5930 - }, - { - "epoch": 8.45, - "learning_rate": 8.613877034929666e-06, - "loss": 0.0799, - "step": 5940 - }, - { - "epoch": 8.46, - "learning_rate": 8.534850640113798e-06, - "loss": 0.0898, - "step": 5950 - }, - { - "epoch": 8.47, - "learning_rate": 8.45582424529793e-06, - "loss": 0.0725, - "step": 5960 - }, - { - "epoch": 8.49, - "learning_rate": 8.376797850482061e-06, - "loss": 0.0835, - "step": 5970 - }, - { - "epoch": 8.5, - "learning_rate": 8.297771455666193e-06, - "loss": 0.1063, - "step": 5980 - }, - { - "epoch": 8.52, - "learning_rate": 8.218745060850325e-06, - "loss": 0.0807, - "step": 5990 - }, - { - "epoch": 8.53, - "learning_rate": 8.139718666034456e-06, - "loss": 0.0726, - "step": 6000 - }, - { - "epoch": 8.55, - "learning_rate": 8.060692271218586e-06, - "loss": 0.0816, - "step": 6010 - }, - { - "epoch": 8.56, - "learning_rate": 7.98166587640272e-06, - "loss": 0.0838, - "step": 6020 - }, - { - "epoch": 8.57, - "learning_rate": 7.90263948158685e-06, - "loss": 0.0752, - "step": 6030 - }, - { - "epoch": 8.59, - "learning_rate": 7.823613086770983e-06, - "loss": 0.09, - "step": 6040 - }, - { - "epoch": 8.6, - "learning_rate": 7.744586691955113e-06, - "loss": 0.1017, - "step": 6050 - }, - { - "epoch": 8.62, - "learning_rate": 7.665560297139245e-06, - "loss": 0.1002, - "step": 6060 - }, - { - "epoch": 8.63, - "learning_rate": 7.586533902323377e-06, - "loss": 0.0853, - "step": 6070 - }, - { - "epoch": 8.65, - "learning_rate": 7.507507507507508e-06, - "loss": 0.0848, - "step": 6080 - }, - { - "epoch": 8.66, - "learning_rate": 7.428481112691639e-06, - "loss": 0.0735, - "step": 6090 - }, - { - "epoch": 8.67, - "learning_rate": 7.3494547178757716e-06, - "loss": 0.0849, - "step": 6100 - }, - { - "epoch": 8.69, - "learning_rate": 7.270428323059902e-06, - "loss": 0.0916, - "step": 6110 - }, - { - "epoch": 8.7, - "learning_rate": 7.191401928244033e-06, - "loss": 0.0903, - "step": 6120 - }, - { - "epoch": 8.72, - "learning_rate": 7.112375533428166e-06, - "loss": 0.0874, - "step": 6130 - }, - { - "epoch": 8.73, - "learning_rate": 7.033349138612297e-06, - "loss": 0.0937, - "step": 6140 - }, - { - "epoch": 8.75, - "learning_rate": 6.9543227437964275e-06, - "loss": 0.089, - "step": 6150 - }, - { - "epoch": 8.76, - "learning_rate": 6.87529634898056e-06, - "loss": 0.0874, - "step": 6160 - }, - { - "epoch": 8.77, - "learning_rate": 6.796269954164691e-06, - "loss": 0.1022, - "step": 6170 - }, - { - "epoch": 8.79, - "learning_rate": 6.7172435593488225e-06, - "loss": 0.0915, - "step": 6180 - }, - { - "epoch": 8.8, - "learning_rate": 6.638217164532955e-06, - "loss": 0.0867, - "step": 6190 - }, - { - "epoch": 8.82, - "learning_rate": 6.559190769717086e-06, - "loss": 0.0794, - "step": 6200 - }, - { - "epoch": 8.83, - "learning_rate": 6.480164374901217e-06, - "loss": 0.0746, - "step": 6210 - }, - { - "epoch": 8.84, - "learning_rate": 6.401137980085349e-06, - "loss": 0.0803, - "step": 6220 - }, - { - "epoch": 8.86, - "learning_rate": 6.32211158526948e-06, - "loss": 0.0726, - "step": 6230 - }, - { - "epoch": 8.87, - "learning_rate": 6.243085190453612e-06, - "loss": 0.0884, - "step": 6240 - }, - { - "epoch": 8.89, - "learning_rate": 6.1640587956377435e-06, - "loss": 0.0933, - "step": 6250 - }, - { - "epoch": 8.9, - "learning_rate": 6.085032400821874e-06, - "loss": 0.0945, - "step": 6260 - }, - { - "epoch": 8.92, - "learning_rate": 6.006006006006006e-06, - "loss": 0.0665, - "step": 6270 - }, - { - "epoch": 8.93, - "learning_rate": 5.926979611190138e-06, - "loss": 0.0775, - "step": 6280 - }, - { - "epoch": 8.94, - "learning_rate": 5.8479532163742686e-06, - "loss": 0.0835, - "step": 6290 - }, - { - "epoch": 8.96, - "learning_rate": 5.7689268215584e-06, - "loss": 0.0954, - "step": 6300 - }, - { - "epoch": 8.97, - "learning_rate": 5.689900426742533e-06, - "loss": 0.0762, - "step": 6310 - }, - { - "epoch": 8.99, - "learning_rate": 5.6108740319266645e-06, - "loss": 0.076, - "step": 6320 - }, - { - "epoch": 9.0, - "eval_accuracy": 0.9874, - "eval_loss": 0.034222181886434555, - "eval_runtime": 59.85, - "eval_samples_per_second": 167.084, - "eval_steps_per_second": 5.23, - "step": 6329 - }, - { - "epoch": 9.0, - "learning_rate": 5.531847637110795e-06, - "loss": 0.066, - "step": 6330 - }, - { - "epoch": 9.02, - "learning_rate": 5.452821242294927e-06, - "loss": 0.0833, - "step": 6340 - }, - { - "epoch": 9.03, - "learning_rate": 5.373794847479059e-06, - "loss": 0.0711, - "step": 6350 - }, - { - "epoch": 9.04, - "learning_rate": 5.2947684526631895e-06, - "loss": 0.0953, - "step": 6360 - }, - { - "epoch": 9.06, - "learning_rate": 5.215742057847321e-06, - "loss": 0.0773, - "step": 6370 - }, - { - "epoch": 9.07, - "learning_rate": 5.136715663031453e-06, - "loss": 0.0961, - "step": 6380 - }, - { - "epoch": 9.09, - "learning_rate": 5.057689268215584e-06, - "loss": 0.0954, - "step": 6390 - }, - { - "epoch": 9.1, - "learning_rate": 4.9786628733997155e-06, + "epoch": 9.96, + "learning_rate": 2.0312817387771686e-07, "loss": 0.0827, - "step": 6400 - }, - { - "epoch": 9.11, - "learning_rate": 4.899636478583847e-06, - "loss": 0.0778, - "step": 6410 - }, - { - "epoch": 9.13, - "learning_rate": 4.820610083767979e-06, - "loss": 0.0818, - "step": 6420 - }, - { - "epoch": 9.14, - "learning_rate": 4.74158368895211e-06, - "loss": 0.0815, - "step": 6430 - }, - { - "epoch": 9.16, - "learning_rate": 4.662557294136241e-06, - "loss": 0.0817, - "step": 6440 - }, - { - "epoch": 9.17, - "learning_rate": 4.583530899320374e-06, - "loss": 0.0855, - "step": 6450 - }, - { - "epoch": 9.19, - "learning_rate": 4.504504504504505e-06, - "loss": 0.0608, - "step": 6460 - }, - { - "epoch": 9.2, - "learning_rate": 4.4254781096886364e-06, - "loss": 0.0625, - "step": 6470 - }, - { - "epoch": 9.21, - "learning_rate": 4.346451714872768e-06, - "loss": 0.0829, - "step": 6480 - }, - { - "epoch": 9.23, - "learning_rate": 4.267425320056899e-06, - "loss": 0.0712, - "step": 6490 - }, - { - "epoch": 9.24, - "learning_rate": 4.188398925241031e-06, - "loss": 0.0912, - "step": 6500 - }, - { - "epoch": 9.26, - "learning_rate": 4.109372530425162e-06, - "loss": 0.0813, - "step": 6510 - }, - { - "epoch": 9.27, - "learning_rate": 4.030346135609293e-06, - "loss": 0.0733, - "step": 6520 - }, - { - "epoch": 9.29, - "learning_rate": 3.951319740793425e-06, - "loss": 0.0747, - "step": 6530 - }, - { - "epoch": 9.3, - "learning_rate": 3.8722933459775566e-06, - "loss": 0.0722, - "step": 6540 - }, - { - "epoch": 9.31, - "learning_rate": 3.7932669511616887e-06, - "loss": 0.0972, - "step": 6550 - }, - { - "epoch": 9.33, - "learning_rate": 3.7142405563458195e-06, - "loss": 0.0794, - "step": 6560 - }, - { - "epoch": 9.34, - "learning_rate": 3.635214161529951e-06, - "loss": 0.0795, - "step": 6570 - }, - { - "epoch": 9.36, - "learning_rate": 3.556187766714083e-06, - "loss": 0.0743, - "step": 6580 - }, - { - "epoch": 9.37, - "learning_rate": 3.4771613718982137e-06, - "loss": 0.0903, - "step": 6590 - }, - { - "epoch": 9.38, - "learning_rate": 3.3981349770823454e-06, - "loss": 0.0802, - "step": 6600 - }, - { - "epoch": 9.4, - "learning_rate": 3.3191085822664775e-06, - "loss": 0.0869, - "step": 6610 - }, - { - "epoch": 9.41, - "learning_rate": 3.2400821874506084e-06, - "loss": 0.0905, - "step": 6620 - }, - { - "epoch": 9.43, - "learning_rate": 3.16105579263474e-06, - "loss": 0.0796, - "step": 6630 - }, - { - "epoch": 9.44, - "learning_rate": 3.0820293978188718e-06, - "loss": 0.0849, - "step": 6640 - }, - { - "epoch": 9.46, - "learning_rate": 3.003003003003003e-06, - "loss": 0.0999, - "step": 6650 - }, - { - "epoch": 9.47, - "learning_rate": 2.9239766081871343e-06, - "loss": 0.085, - "step": 6660 - }, - { - "epoch": 9.48, - "learning_rate": 2.8449502133712664e-06, - "loss": 0.0716, - "step": 6670 - }, - { - "epoch": 9.5, - "learning_rate": 2.7659238185553977e-06, - "loss": 0.083, - "step": 6680 - }, - { - "epoch": 9.51, - "learning_rate": 2.6868974237395294e-06, - "loss": 0.0889, - "step": 6690 - }, - { - "epoch": 9.53, - "learning_rate": 2.6078710289236606e-06, - "loss": 0.0897, - "step": 6700 - }, - { - "epoch": 9.54, - "learning_rate": 2.528844634107792e-06, - "loss": 0.0854, - "step": 6710 - }, - { - "epoch": 9.56, - "learning_rate": 2.4498182392919236e-06, - "loss": 0.0946, - "step": 6720 - }, - { - "epoch": 9.57, - "learning_rate": 2.370791844476055e-06, - "loss": 0.0767, - "step": 6730 - }, - { - "epoch": 9.58, - "learning_rate": 2.291765449660187e-06, - "loss": 0.0822, - "step": 6740 - }, - { - "epoch": 9.6, - "learning_rate": 2.2127390548443182e-06, - "loss": 0.0847, - "step": 6750 - }, - { - "epoch": 9.61, - "learning_rate": 2.1337126600284495e-06, - "loss": 0.085, - "step": 6760 - }, - { - "epoch": 9.63, - "learning_rate": 2.054686265212581e-06, - "loss": 0.0721, - "step": 6770 - }, - { - "epoch": 9.64, - "learning_rate": 1.9756598703967124e-06, - "loss": 0.0767, - "step": 6780 - }, - { - "epoch": 9.66, - "learning_rate": 1.8966334755808443e-06, - "loss": 0.0771, - "step": 6790 - }, - { - "epoch": 9.67, - "learning_rate": 1.8176070807649756e-06, - "loss": 0.0869, - "step": 6800 - }, - { - "epoch": 9.68, - "learning_rate": 1.7385806859491069e-06, - "loss": 0.0791, - "step": 6810 - }, - { - "epoch": 9.7, - "learning_rate": 1.6595542911332388e-06, - "loss": 0.0707, - "step": 6820 - }, - { - "epoch": 9.71, - "learning_rate": 1.58052789631737e-06, - "loss": 0.0617, - "step": 6830 - }, - { - "epoch": 9.73, - "learning_rate": 1.5015015015015015e-06, - "loss": 0.0894, - "step": 6840 - }, - { - "epoch": 9.74, - "learning_rate": 1.4224751066856332e-06, - "loss": 0.0786, - "step": 6850 - }, - { - "epoch": 9.75, - "learning_rate": 1.3434487118697647e-06, - "loss": 0.0733, - "step": 6860 - }, - { - "epoch": 9.77, - "learning_rate": 1.264422317053896e-06, - "loss": 0.0771, - "step": 6870 - }, - { - "epoch": 9.78, - "learning_rate": 1.1853959222380274e-06, - "loss": 0.0838, - "step": 6880 - }, - { - "epoch": 9.8, - "learning_rate": 1.1063695274221591e-06, - "loss": 0.0855, - "step": 6890 - }, - { - "epoch": 9.81, - "learning_rate": 1.0273431326062906e-06, - "loss": 0.0696, - "step": 6900 - }, - { - "epoch": 9.83, - "learning_rate": 9.483167377904222e-07, - "loss": 0.0867, - "step": 6910 - }, - { - "epoch": 9.84, - "learning_rate": 8.692903429745534e-07, - "loss": 0.0822, - "step": 6920 - }, - { - "epoch": 9.85, - "learning_rate": 7.90263948158685e-07, - "loss": 0.0768, - "step": 6930 - }, - { - "epoch": 9.87, - "learning_rate": 7.112375533428166e-07, - "loss": 0.0716, - "step": 6940 - }, - { - "epoch": 9.88, - "learning_rate": 6.32211158526948e-07, - "loss": 0.0722, - "step": 6950 - }, - { - "epoch": 9.9, - "learning_rate": 5.531847637110796e-07, - "loss": 0.0594, - "step": 6960 - }, - { - "epoch": 9.91, - "learning_rate": 4.741583688952111e-07, - "loss": 0.0749, - "step": 6970 - }, - { - "epoch": 9.93, - "learning_rate": 3.951319740793425e-07, - "loss": 0.0972, - "step": 6980 - }, - { - "epoch": 9.94, - "learning_rate": 3.16105579263474e-07, - "loss": 0.0951, - "step": 6990 - }, - { - "epoch": 9.95, - "learning_rate": 2.3707918444760554e-07, - "loss": 0.0805, - "step": 7000 - }, - { - "epoch": 9.97, - "learning_rate": 1.58052789631737e-07, - "loss": 0.08, - "step": 7010 + "step": 5450 }, { "epoch": 9.98, - "learning_rate": 7.90263948158685e-08, - "loss": 0.0925, - "step": 7020 + "learning_rate": 1.0156408693885843e-07, + "loss": 0.1004, + "step": 5460 }, { "epoch": 10.0, "learning_rate": 0.0, - "loss": 0.0887, - "step": 7030 + "loss": 0.0778, + "step": 5470 }, { "epoch": 10.0, - "eval_accuracy": 0.9844, - "eval_loss": 0.0412600003182888, - "eval_runtime": 59.4728, - "eval_samples_per_second": 168.144, - "eval_steps_per_second": 5.263, - "step": 7030 + "eval_accuracy": 0.9807, + "eval_loss": 0.05392194166779518, + "eval_runtime": 108.2183, + "eval_samples_per_second": 277.217, + "eval_steps_per_second": 8.668, + "step": 5470 }, { "epoch": 10.0, - "step": 7030, - "total_flos": 2.23628338666187e+19, - "train_loss": 0.1441501469903786, - "train_runtime": 13283.8875, - "train_samples_per_second": 67.751, - "train_steps_per_second": 0.529 + "step": 5470, + "total_flos": 1.73991922900992e+19, + "train_loss": 0.15263754843139998, + "train_runtime": 7208.4995, + "train_samples_per_second": 97.108, + "train_steps_per_second": 0.759 } ], - "max_steps": 7030, + "max_steps": 5470, "num_train_epochs": 10, - "total_flos": 2.23628338666187e+19, + "total_flos": 1.73991922900992e+19, "trial_name": null, "trial_params": null }