{ "best_metric": null, "best_model_checkpoint": null, "epoch": 1.4438799076212472, "eval_steps": 1000, "global_step": 1922, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 1.3824884792626728e-07, "loss": 0.7351, "step": 1 }, { "epoch": 0.0, "learning_rate": 1.3824884792626729e-06, "loss": 0.7455, "step": 10 }, { "epoch": 0.01, "learning_rate": 2.7649769585253458e-06, "loss": 0.7061, "step": 20 }, { "epoch": 0.01, "learning_rate": 4.147465437788019e-06, "loss": 0.6593, "step": 30 }, { "epoch": 0.02, "learning_rate": 5.5299539170506915e-06, "loss": 0.6177, "step": 40 }, { "epoch": 0.02, "learning_rate": 6.912442396313364e-06, "loss": 0.5817, "step": 50 }, { "epoch": 0.03, "learning_rate": 8.294930875576038e-06, "loss": 0.5095, "step": 60 }, { "epoch": 0.03, "learning_rate": 9.67741935483871e-06, "loss": 0.4615, "step": 70 }, { "epoch": 0.04, "learning_rate": 1.1059907834101383e-05, "loss": 0.4126, "step": 80 }, { "epoch": 0.04, "learning_rate": 1.2442396313364056e-05, "loss": 0.4067, "step": 90 }, { "epoch": 0.05, "learning_rate": 1.3824884792626728e-05, "loss": 0.3974, "step": 100 }, { "epoch": 0.05, "learning_rate": 1.5207373271889403e-05, "loss": 0.3864, "step": 110 }, { "epoch": 0.06, "learning_rate": 1.6589861751152075e-05, "loss": 0.4016, "step": 120 }, { "epoch": 0.06, "learning_rate": 1.7972350230414745e-05, "loss": 0.3873, "step": 130 }, { "epoch": 0.06, "learning_rate": 1.935483870967742e-05, "loss": 0.3837, "step": 140 }, { "epoch": 0.07, "learning_rate": 2.0737327188940094e-05, "loss": 0.3868, "step": 150 }, { "epoch": 0.07, "learning_rate": 2.2119815668202766e-05, "loss": 0.3817, "step": 160 }, { "epoch": 0.08, "learning_rate": 2.350230414746544e-05, "loss": 0.3593, "step": 170 }, { "epoch": 0.08, "learning_rate": 2.488479262672811e-05, "loss": 0.3641, "step": 180 }, { "epoch": 0.09, "learning_rate": 2.6267281105990784e-05, "loss": 0.3497, "step": 190 }, { "epoch": 0.09, "learning_rate": 2.7649769585253457e-05, "loss": 0.3679, "step": 200 }, { "epoch": 0.1, "learning_rate": 2.903225806451613e-05, "loss": 0.3751, "step": 210 }, { "epoch": 0.1, "learning_rate": 2.9999960619075335e-05, "loss": 0.3828, "step": 220 }, { "epoch": 0.11, "learning_rate": 2.9999260519500367e-05, "loss": 0.3763, "step": 230 }, { "epoch": 0.11, "learning_rate": 2.9997685335280646e-05, "loss": 0.3553, "step": 240 }, { "epoch": 0.12, "learning_rate": 2.9995235158315353e-05, "loss": 0.3589, "step": 250 }, { "epoch": 0.12, "learning_rate": 2.999191013155234e-05, "loss": 0.3585, "step": 260 }, { "epoch": 0.12, "learning_rate": 2.998771044897983e-05, "loss": 0.3529, "step": 270 }, { "epoch": 0.13, "learning_rate": 2.9982636355615092e-05, "loss": 0.3303, "step": 280 }, { "epoch": 0.13, "learning_rate": 2.997668814749012e-05, "loss": 0.3696, "step": 290 }, { "epoch": 0.14, "learning_rate": 2.99698661716344e-05, "loss": 0.3353, "step": 300 }, { "epoch": 0.14, "learning_rate": 2.9962170826054645e-05, "loss": 0.3562, "step": 310 }, { "epoch": 0.15, "learning_rate": 2.995360255971157e-05, "loss": 0.3652, "step": 320 }, { "epoch": 0.15, "learning_rate": 2.994416187249371e-05, "loss": 0.3522, "step": 330 }, { "epoch": 0.16, "learning_rate": 2.9933849315188233e-05, "loss": 0.3754, "step": 340 }, { "epoch": 0.16, "learning_rate": 2.992266548944885e-05, "loss": 0.3348, "step": 350 }, { "epoch": 0.17, "learning_rate": 2.991061104776067e-05, "loss": 0.3513, "step": 360 }, { "epoch": 0.17, "learning_rate": 2.9897686693402138e-05, "loss": 0.3214, "step": 370 }, { "epoch": 0.18, "learning_rate": 2.9883893180404046e-05, "loss": 0.3451, "step": 380 }, { "epoch": 0.18, "learning_rate": 2.986923131350549e-05, "loss": 0.3579, "step": 390 }, { "epoch": 0.18, "learning_rate": 2.9853701948106944e-05, "loss": 0.3353, "step": 400 }, { "epoch": 0.19, "learning_rate": 2.9837305990220357e-05, "loss": 0.3374, "step": 410 }, { "epoch": 0.19, "learning_rate": 2.982004439641628e-05, "loss": 0.3544, "step": 420 }, { "epoch": 0.2, "learning_rate": 2.980191817376808e-05, "loss": 0.3313, "step": 430 }, { "epoch": 0.2, "learning_rate": 2.9782928379793154e-05, "loss": 0.3488, "step": 440 }, { "epoch": 0.21, "learning_rate": 2.976307612239127e-05, "loss": 0.3384, "step": 450 }, { "epoch": 0.21, "learning_rate": 2.97423625597799e-05, "loss": 0.3359, "step": 460 }, { "epoch": 0.22, "learning_rate": 2.9720788900426657e-05, "loss": 0.3353, "step": 470 }, { "epoch": 0.22, "learning_rate": 2.969835640297879e-05, "loss": 0.356, "step": 480 }, { "epoch": 0.23, "learning_rate": 2.967506637618976e-05, "loss": 0.3575, "step": 490 }, { "epoch": 0.23, "learning_rate": 2.9650920178842874e-05, "loss": 0.3163, "step": 500 }, { "epoch": 0.24, "learning_rate": 2.9625919219672017e-05, "loss": 0.3277, "step": 510 }, { "epoch": 0.24, "learning_rate": 2.960006495727946e-05, "loss": 0.3627, "step": 520 }, { "epoch": 0.24, "learning_rate": 2.9573358900050764e-05, "loss": 0.3412, "step": 530 }, { "epoch": 0.25, "learning_rate": 2.9545802606066778e-05, "loss": 0.3464, "step": 540 }, { "epoch": 0.25, "learning_rate": 2.9517397683012747e-05, "loss": 0.3417, "step": 550 }, { "epoch": 0.26, "learning_rate": 2.9488145788084502e-05, "loss": 0.349, "step": 560 }, { "epoch": 0.26, "learning_rate": 2.945804862789178e-05, "loss": 0.3497, "step": 570 }, { "epoch": 0.27, "learning_rate": 2.942710795835866e-05, "loss": 0.3318, "step": 580 }, { "epoch": 0.27, "learning_rate": 2.9395325584621122e-05, "loss": 0.3249, "step": 590 }, { "epoch": 0.28, "learning_rate": 2.9362703360921722e-05, "loss": 0.3346, "step": 600 }, { "epoch": 0.28, "learning_rate": 2.932924319050143e-05, "loss": 0.3401, "step": 610 }, { "epoch": 0.29, "learning_rate": 2.9294947025488568e-05, "loss": 0.327, "step": 620 }, { "epoch": 0.29, "learning_rate": 2.925981686678494e-05, "loss": 0.3185, "step": 630 }, { "epoch": 0.3, "learning_rate": 2.9223854763949082e-05, "loss": 0.3411, "step": 640 }, { "epoch": 0.3, "learning_rate": 2.9187062815076688e-05, "loss": 0.3401, "step": 650 }, { "epoch": 0.3, "learning_rate": 2.914944316667822e-05, "loss": 0.3443, "step": 660 }, { "epoch": 0.31, "learning_rate": 2.9110998013553653e-05, "loss": 0.3336, "step": 670 }, { "epoch": 0.31, "learning_rate": 2.9071729598664433e-05, "loss": 0.3271, "step": 680 }, { "epoch": 0.32, "learning_rate": 2.9031640213002638e-05, "loss": 0.3233, "step": 690 }, { "epoch": 0.32, "learning_rate": 2.899073219545729e-05, "loss": 0.3548, "step": 700 }, { "epoch": 0.33, "learning_rate": 2.8949007932677915e-05, "loss": 0.3243, "step": 710 }, { "epoch": 0.33, "learning_rate": 2.89064698589353e-05, "loss": 0.3151, "step": 720 }, { "epoch": 0.34, "learning_rate": 2.8863120455979458e-05, "loss": 0.325, "step": 730 }, { "epoch": 0.34, "learning_rate": 2.8818962252894872e-05, "loss": 0.3406, "step": 740 }, { "epoch": 0.35, "learning_rate": 2.8773997825952914e-05, "loss": 0.3059, "step": 750 }, { "epoch": 0.35, "learning_rate": 2.872822979846154e-05, "loss": 0.3236, "step": 760 }, { "epoch": 0.36, "learning_rate": 2.8681660840612262e-05, "loss": 0.3465, "step": 770 }, { "epoch": 0.36, "learning_rate": 2.8634293669324353e-05, "loss": 0.34, "step": 780 }, { "epoch": 0.36, "learning_rate": 2.8586131048086334e-05, "loss": 0.3319, "step": 790 }, { "epoch": 0.37, "learning_rate": 2.853717578679474e-05, "loss": 0.3322, "step": 800 }, { "epoch": 0.37, "learning_rate": 2.848743074159021e-05, "loss": 0.3125, "step": 810 }, { "epoch": 0.38, "learning_rate": 2.8436898814690837e-05, "loss": 0.3411, "step": 820 }, { "epoch": 0.38, "learning_rate": 2.838558295422284e-05, "loss": 0.3371, "step": 830 }, { "epoch": 0.39, "learning_rate": 2.833348615404859e-05, "loss": 0.3336, "step": 840 }, { "epoch": 0.39, "learning_rate": 2.8280611453591908e-05, "loss": 0.3335, "step": 850 }, { "epoch": 0.4, "learning_rate": 2.8226961937660773e-05, "loss": 0.3131, "step": 860 }, { "epoch": 0.4, "learning_rate": 2.817254073626733e-05, "loss": 0.3202, "step": 870 }, { "epoch": 0.41, "learning_rate": 2.811735102444528e-05, "loss": 0.3265, "step": 880 }, { "epoch": 0.41, "learning_rate": 2.8061396022064657e-05, "loss": 0.3162, "step": 890 }, { "epoch": 0.42, "learning_rate": 2.8004678993643952e-05, "loss": 0.3124, "step": 900 }, { "epoch": 0.42, "learning_rate": 2.7947203248159665e-05, "loss": 0.3289, "step": 910 }, { "epoch": 0.42, "learning_rate": 2.788897213885327e-05, "loss": 0.3242, "step": 920 }, { "epoch": 0.43, "learning_rate": 2.782998906303555e-05, "loss": 0.3038, "step": 930 }, { "epoch": 0.43, "learning_rate": 2.777025746188842e-05, "loss": 0.3319, "step": 940 }, { "epoch": 0.44, "learning_rate": 2.7709780820264147e-05, "loss": 0.3372, "step": 950 }, { "epoch": 0.44, "learning_rate": 2.764856266648202e-05, "loss": 0.306, "step": 960 }, { "epoch": 1.0, "learning_rate": 2.758660657212255e-05, "loss": 0.3001, "step": 970 }, { "epoch": 1.01, "learning_rate": 2.7523916151819048e-05, "loss": 0.3044, "step": 980 }, { "epoch": 1.01, "learning_rate": 2.746049506304678e-05, "loss": 0.3114, "step": 990 }, { "epoch": 1.02, "learning_rate": 2.7396347005909535e-05, "loss": 0.2982, "step": 1000 }, { "epoch": 1.02, "eval_loss": 0.44782933592796326, "eval_runtime": 6.8987, "eval_samples_per_second": 20.294, "eval_steps_per_second": 5.073, "step": 1000 }, { "epoch": 1.02, "learning_rate": 2.733147572292381e-05, "loss": 0.3091, "step": 1010 }, { "epoch": 1.03, "learning_rate": 2.7265884998800434e-05, "loss": 0.3011, "step": 1020 }, { "epoch": 1.03, "learning_rate": 2.7199578660223743e-05, "loss": 0.2794, "step": 1030 }, { "epoch": 1.04, "learning_rate": 2.7132560575628377e-05, "loss": 0.2956, "step": 1040 }, { "epoch": 1.04, "learning_rate": 2.7064834654973534e-05, "loss": 0.3098, "step": 1050 }, { "epoch": 1.05, "learning_rate": 2.6996404849514885e-05, "loss": 0.3315, "step": 1060 }, { "epoch": 1.05, "learning_rate": 2.6927275151574053e-05, "loss": 0.305, "step": 1070 }, { "epoch": 1.05, "learning_rate": 2.6857449594305674e-05, "loss": 0.2986, "step": 1080 }, { "epoch": 1.06, "learning_rate": 2.678693225146211e-05, "loss": 0.269, "step": 1090 }, { "epoch": 1.06, "learning_rate": 2.6715727237155777e-05, "loss": 0.3124, "step": 1100 }, { "epoch": 1.07, "learning_rate": 2.6643838705619117e-05, "loss": 0.303, "step": 1110 }, { "epoch": 1.07, "learning_rate": 2.6571270850962234e-05, "loss": 0.3084, "step": 1120 }, { "epoch": 1.08, "learning_rate": 2.6498027906928195e-05, "loss": 0.31, "step": 1130 }, { "epoch": 1.08, "learning_rate": 2.6424114146646043e-05, "loss": 0.3203, "step": 1140 }, { "epoch": 1.09, "learning_rate": 2.6349533882381475e-05, "loss": 0.2963, "step": 1150 }, { "epoch": 1.09, "learning_rate": 2.6274291465285266e-05, "loss": 0.3042, "step": 1160 }, { "epoch": 1.1, "learning_rate": 2.6198391285139417e-05, "loss": 0.314, "step": 1170 }, { "epoch": 1.1, "learning_rate": 2.612183777010104e-05, "loss": 0.3144, "step": 1180 }, { "epoch": 1.11, "learning_rate": 2.6044635386444024e-05, "loss": 0.3157, "step": 1190 }, { "epoch": 1.11, "learning_rate": 2.5966788638298443e-05, "loss": 0.2958, "step": 1200 }, { "epoch": 1.12, "learning_rate": 2.5888302067387793e-05, "loss": 0.2867, "step": 1210 }, { "epoch": 1.12, "learning_rate": 2.5809180252764022e-05, "loss": 0.2999, "step": 1220 }, { "epoch": 1.12, "learning_rate": 2.572942781054036e-05, "loss": 0.2888, "step": 1230 }, { "epoch": 1.13, "learning_rate": 2.564904939362204e-05, "loss": 0.3036, "step": 1240 }, { "epoch": 1.13, "learning_rate": 2.5568049691434794e-05, "loss": 0.3127, "step": 1250 }, { "epoch": 1.14, "learning_rate": 2.5486433429651304e-05, "loss": 0.2939, "step": 1260 }, { "epoch": 1.14, "learning_rate": 2.5404205369915473e-05, "loss": 0.3096, "step": 1270 }, { "epoch": 1.15, "learning_rate": 2.532137030956464e-05, "loss": 0.3039, "step": 1280 }, { "epoch": 1.15, "learning_rate": 2.523793308134967e-05, "loss": 0.2948, "step": 1290 }, { "epoch": 1.16, "learning_rate": 2.5153898553153024e-05, "loss": 0.3053, "step": 1300 }, { "epoch": 1.16, "learning_rate": 2.506927162770475e-05, "loss": 0.2962, "step": 1310 }, { "epoch": 1.17, "learning_rate": 2.4984057242296464e-05, "loss": 0.3027, "step": 1320 }, { "epoch": 1.17, "learning_rate": 2.489826036849325e-05, "loss": 0.3051, "step": 1330 }, { "epoch": 1.18, "learning_rate": 2.4811886011843673e-05, "loss": 0.3214, "step": 1340 }, { "epoch": 1.18, "learning_rate": 2.4724939211587706e-05, "loss": 0.3002, "step": 1350 }, { "epoch": 1.18, "learning_rate": 2.4637425040362744e-05, "loss": 0.3177, "step": 1360 }, { "epoch": 1.19, "learning_rate": 2.4549348603907658e-05, "loss": 0.3141, "step": 1370 }, { "epoch": 1.19, "learning_rate": 2.4460715040764916e-05, "loss": 0.3191, "step": 1380 }, { "epoch": 1.2, "learning_rate": 2.4371529521980775e-05, "loss": 0.2736, "step": 1390 }, { "epoch": 1.2, "learning_rate": 2.428179725080362e-05, "loss": 0.3063, "step": 1400 }, { "epoch": 1.21, "learning_rate": 2.419152346238038e-05, "loss": 0.2965, "step": 1410 }, { "epoch": 1.21, "learning_rate": 2.410071342345111e-05, "loss": 0.2942, "step": 1420 }, { "epoch": 1.22, "learning_rate": 2.4009372432041702e-05, "loss": 0.2895, "step": 1430 }, { "epoch": 1.22, "learning_rate": 2.3917505817154795e-05, "loss": 0.3084, "step": 1440 }, { "epoch": 1.23, "learning_rate": 2.3825118938458894e-05, "loss": 0.2849, "step": 1450 }, { "epoch": 1.23, "learning_rate": 2.373221718597564e-05, "loss": 0.2923, "step": 1460 }, { "epoch": 1.24, "learning_rate": 2.3638805979765387e-05, "loss": 0.3062, "step": 1470 }, { "epoch": 1.24, "learning_rate": 2.3544890769610936e-05, "loss": 0.2832, "step": 1480 }, { "epoch": 1.24, "learning_rate": 2.3450477034699632e-05, "loss": 0.2823, "step": 1490 }, { "epoch": 1.25, "learning_rate": 2.335557028330366e-05, "loss": 0.2822, "step": 1500 }, { "epoch": 1.25, "learning_rate": 2.326017605245872e-05, "loss": 0.2885, "step": 1510 }, { "epoch": 1.26, "learning_rate": 2.3164299907640955e-05, "loss": 0.2852, "step": 1520 }, { "epoch": 1.26, "learning_rate": 2.3067947442442264e-05, "loss": 0.3022, "step": 1530 }, { "epoch": 1.27, "learning_rate": 2.2971124278243957e-05, "loss": 0.2666, "step": 1540 }, { "epoch": 1.27, "learning_rate": 2.28738360638888e-05, "loss": 0.2892, "step": 1550 }, { "epoch": 1.28, "learning_rate": 2.2776088475351445e-05, "loss": 0.2689, "step": 1560 }, { "epoch": 1.28, "learning_rate": 2.2677887215407278e-05, "loss": 0.2864, "step": 1570 }, { "epoch": 1.29, "learning_rate": 2.257923801329973e-05, "loss": 0.281, "step": 1580 }, { "epoch": 1.29, "learning_rate": 2.248014662440599e-05, "loss": 0.3068, "step": 1590 }, { "epoch": 1.3, "learning_rate": 2.238061882990126e-05, "loss": 0.2753, "step": 1600 }, { "epoch": 1.3, "learning_rate": 2.2280660436421443e-05, "loss": 0.2701, "step": 1610 }, { "epoch": 1.3, "learning_rate": 2.2180277275724385e-05, "loss": 0.2891, "step": 1620 }, { "epoch": 1.31, "learning_rate": 2.2079475204349645e-05, "loss": 0.2691, "step": 1630 }, { "epoch": 1.31, "learning_rate": 2.1978260103276796e-05, "loss": 0.2969, "step": 1640 }, { "epoch": 1.32, "learning_rate": 2.187663787758234e-05, "loss": 0.2629, "step": 1650 }, { "epoch": 1.32, "learning_rate": 2.177461445609518e-05, "loss": 0.2746, "step": 1660 }, { "epoch": 1.33, "learning_rate": 2.1672195791050712e-05, "loss": 0.2775, "step": 1670 }, { "epoch": 1.33, "learning_rate": 2.1569387857743596e-05, "loss": 0.2654, "step": 1680 }, { "epoch": 1.34, "learning_rate": 2.1466196654179107e-05, "loss": 0.2678, "step": 1690 }, { "epoch": 1.34, "learning_rate": 2.1362628200723228e-05, "loss": 0.2606, "step": 1700 }, { "epoch": 1.35, "learning_rate": 2.1258688539751387e-05, "loss": 0.2886, "step": 1710 }, { "epoch": 1.35, "learning_rate": 2.115438373529596e-05, "loss": 0.2799, "step": 1720 }, { "epoch": 1.36, "learning_rate": 2.104971987269245e-05, "loss": 0.2804, "step": 1730 }, { "epoch": 1.36, "learning_rate": 2.0944703058224504e-05, "loss": 0.2583, "step": 1740 }, { "epoch": 1.36, "learning_rate": 2.0839339418767616e-05, "loss": 0.2857, "step": 1750 }, { "epoch": 1.37, "learning_rate": 2.0733635101431694e-05, "loss": 0.2692, "step": 1760 }, { "epoch": 1.37, "learning_rate": 2.0627596273202435e-05, "loss": 0.2759, "step": 1770 }, { "epoch": 1.38, "learning_rate": 2.05212291205815e-05, "loss": 0.2685, "step": 1780 }, { "epoch": 1.38, "learning_rate": 2.0414539849225637e-05, "loss": 0.2511, "step": 1790 }, { "epoch": 1.39, "learning_rate": 2.0307534683584565e-05, "loss": 0.2764, "step": 1800 }, { "epoch": 1.39, "learning_rate": 2.0200219866537882e-05, "loss": 0.2763, "step": 1810 }, { "epoch": 1.4, "learning_rate": 2.0092601659030807e-05, "loss": 0.2731, "step": 1820 }, { "epoch": 1.4, "learning_rate": 1.9984686339708927e-05, "loss": 0.2803, "step": 1830 }, { "epoch": 1.41, "learning_rate": 1.9876480204551894e-05, "loss": 0.281, "step": 1840 }, { "epoch": 1.41, "learning_rate": 1.976798956650607e-05, "loss": 0.2802, "step": 1850 }, { "epoch": 1.42, "learning_rate": 1.9659220755116277e-05, "loss": 0.2743, "step": 1860 }, { "epoch": 1.42, "learning_rate": 1.9550180116156447e-05, "loss": 0.2579, "step": 1870 }, { "epoch": 1.42, "learning_rate": 1.9440874011259458e-05, "loss": 0.2721, "step": 1880 }, { "epoch": 1.43, "learning_rate": 1.9331308817545963e-05, "loss": 0.2713, "step": 1890 }, { "epoch": 1.43, "learning_rate": 1.922149092725233e-05, "loss": 0.2874, "step": 1900 }, { "epoch": 1.44, "learning_rate": 1.911142674735771e-05, "loss": 0.2766, "step": 1910 }, { "epoch": 1.44, "learning_rate": 1.900112269921026e-05, "loss": 0.2695, "step": 1920 }, { "epoch": 1.44, "step": 1922, "total_flos": 3.4599459174219776e+17, "train_loss": 0.3275021193364904, "train_runtime": 3966.2406, "train_samples_per_second": 4.365, "train_steps_per_second": 1.092 } ], "logging_steps": 10, "max_steps": 4330, "num_train_epochs": 2, "save_steps": 1000, "total_flos": 3.4599459174219776e+17, "trial_name": null, "trial_params": null }