{ "best_metric": null, "best_model_checkpoint": null, "epoch": 26.0, "global_step": 95628, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 5e-09, "loss": 10.5649, "step": 1 }, { "epoch": 0.14, "learning_rate": 2.5e-06, "loss": 9.6075, "step": 500 }, { "epoch": 0.27, "learning_rate": 5e-06, "loss": 8.0262, "step": 1000 }, { "epoch": 0.41, "learning_rate": 7.5e-06, "loss": 6.9785, "step": 1500 }, { "epoch": 0.54, "learning_rate": 1e-05, "loss": 6.6731, "step": 2000 }, { "epoch": 0.68, "learning_rate": 1.25e-05, "loss": 6.5046, "step": 2500 }, { "epoch": 0.82, "learning_rate": 1.5e-05, "loss": 6.3911, "step": 3000 }, { "epoch": 0.95, "learning_rate": 1.75e-05, "loss": 6.2957, "step": 3500 }, { "epoch": 1.09, "learning_rate": 2e-05, "loss": 6.2241, "step": 4000 }, { "epoch": 1.22, "learning_rate": 2.25e-05, "loss": 6.1595, "step": 4500 }, { "epoch": 1.36, "learning_rate": 2.5e-05, "loss": 6.1063, "step": 5000 }, { "epoch": 1.5, "learning_rate": 2.7500000000000004e-05, "loss": 6.0549, "step": 5500 }, { "epoch": 1.63, "learning_rate": 3e-05, "loss": 6.0125, "step": 6000 }, { "epoch": 1.77, "learning_rate": 3.2500000000000004e-05, "loss": 5.9775, "step": 6500 }, { "epoch": 1.9, "learning_rate": 3.5e-05, "loss": 5.9469, "step": 7000 }, { "epoch": 2.04, "learning_rate": 3.7500000000000003e-05, "loss": 5.9122, "step": 7500 }, { "epoch": 2.18, "learning_rate": 4e-05, "loss": 5.8878, "step": 8000 }, { "epoch": 2.31, "learning_rate": 4.2495e-05, "loss": 5.8635, "step": 8500 }, { "epoch": 2.45, "learning_rate": 4.4995000000000005e-05, "loss": 5.838, "step": 9000 }, { "epoch": 2.58, "learning_rate": 4.7495e-05, "loss": 5.8177, "step": 9500 }, { "epoch": 2.72, "learning_rate": 4.9995000000000005e-05, "loss": 5.8012, "step": 10000 }, { "epoch": 2.85, "learning_rate": 4.99829592116069e-05, "loss": 5.7852, "step": 10500 }, { "epoch": 2.99, "learning_rate": 4.9965849986312624e-05, "loss": 5.768, "step": 11000 }, { "epoch": 3.13, "learning_rate": 4.9948740761018345e-05, "loss": 5.7555, "step": 11500 }, { "epoch": 3.26, "learning_rate": 4.9931631535724066e-05, "loss": 5.742, "step": 12000 }, { "epoch": 3.4, "learning_rate": 4.991452231042979e-05, "loss": 5.728, "step": 12500 }, { "epoch": 3.53, "learning_rate": 4.98974473035861e-05, "loss": 5.7197, "step": 13000 }, { "epoch": 3.67, "learning_rate": 4.988033807829182e-05, "loss": 5.7113, "step": 13500 }, { "epoch": 3.81, "learning_rate": 4.986322885299754e-05, "loss": 5.7034, "step": 14000 }, { "epoch": 3.94, "learning_rate": 4.984611962770326e-05, "loss": 5.6963, "step": 14500 }, { "epoch": 4.08, "learning_rate": 4.982904462085957e-05, "loss": 5.6869, "step": 15000 }, { "epoch": 4.21, "learning_rate": 4.981193539556529e-05, "loss": 5.6802, "step": 15500 }, { "epoch": 4.35, "learning_rate": 4.9794826170271015e-05, "loss": 5.6739, "step": 16000 }, { "epoch": 4.49, "learning_rate": 4.977771694497673e-05, "loss": 5.6674, "step": 16500 }, { "epoch": 4.62, "learning_rate": 4.9760641938133046e-05, "loss": 5.6611, "step": 17000 }, { "epoch": 4.76, "learning_rate": 4.974353271283876e-05, "loss": 5.6567, "step": 17500 }, { "epoch": 4.89, "learning_rate": 4.972642348754449e-05, "loss": 5.6491, "step": 18000 }, { "epoch": 5.03, "learning_rate": 4.97093142622502e-05, "loss": 5.6447, "step": 18500 }, { "epoch": 5.17, "learning_rate": 4.969223925540652e-05, "loss": 5.6407, "step": 19000 }, { "epoch": 5.3, "learning_rate": 4.9675130030112235e-05, "loss": 5.6356, "step": 19500 }, { "epoch": 5.44, "learning_rate": 4.965802080481796e-05, "loss": 5.6318, "step": 20000 }, { "epoch": 5.57, "learning_rate": 4.964091157952368e-05, "loss": 5.6285, "step": 20500 }, { "epoch": 5.71, "learning_rate": 4.9623802354229405e-05, "loss": 5.6227, "step": 21000 }, { "epoch": 5.85, "learning_rate": 4.960672734738571e-05, "loss": 5.6178, "step": 21500 }, { "epoch": 5.98, "learning_rate": 4.958961812209144e-05, "loss": 5.6137, "step": 22000 }, { "epoch": 6.12, "learning_rate": 4.957250889679715e-05, "loss": 5.6102, "step": 22500 }, { "epoch": 6.25, "learning_rate": 4.955539967150288e-05, "loss": 5.6042, "step": 23000 }, { "epoch": 6.39, "learning_rate": 4.953832466465918e-05, "loss": 5.6023, "step": 23500 }, { "epoch": 6.53, "learning_rate": 4.952121543936491e-05, "loss": 5.6011, "step": 24000 }, { "epoch": 6.66, "learning_rate": 4.9504106214070626e-05, "loss": 5.5958, "step": 24500 }, { "epoch": 6.8, "learning_rate": 4.9486996988776353e-05, "loss": 5.5914, "step": 25000 }, { "epoch": 6.93, "learning_rate": 4.946992198193266e-05, "loss": 5.5901, "step": 25500 }, { "epoch": 7.07, "learning_rate": 4.9452812756638385e-05, "loss": 5.5889, "step": 26000 }, { "epoch": 7.21, "learning_rate": 4.94357035313441e-05, "loss": 5.5859, "step": 26500 }, { "epoch": 7.34, "learning_rate": 4.941862852450042e-05, "loss": 5.5826, "step": 27000 }, { "epoch": 7.48, "learning_rate": 4.940151929920613e-05, "loss": 5.5774, "step": 27500 }, { "epoch": 7.61, "learning_rate": 4.938441007391186e-05, "loss": 5.5748, "step": 28000 }, { "epoch": 7.75, "learning_rate": 4.9367300848617574e-05, "loss": 5.5737, "step": 28500 }, { "epoch": 7.88, "learning_rate": 4.93501916233233e-05, "loss": 5.5726, "step": 29000 }, { "epoch": 8.02, "learning_rate": 4.9333082398029016e-05, "loss": 5.5696, "step": 29500 }, { "epoch": 8.16, "learning_rate": 4.9315973172734744e-05, "loss": 5.5651, "step": 30000 }, { "epoch": 8.29, "learning_rate": 4.929886394744046e-05, "loss": 5.562, "step": 30500 }, { "epoch": 8.43, "learning_rate": 4.9281788940596776e-05, "loss": 5.5569, "step": 31000 }, { "epoch": 8.56, "learning_rate": 4.926467971530249e-05, "loss": 5.5546, "step": 31500 }, { "epoch": 8.7, "learning_rate": 4.924760470845881e-05, "loss": 5.5564, "step": 32000 }, { "epoch": 8.84, "learning_rate": 4.923049548316452e-05, "loss": 5.5569, "step": 32500 }, { "epoch": 8.97, "learning_rate": 4.921338625787025e-05, "loss": 5.5486, "step": 33000 }, { "epoch": 9.11, "learning_rate": 4.9196277032575965e-05, "loss": 5.5491, "step": 33500 }, { "epoch": 9.24, "learning_rate": 4.917916780728169e-05, "loss": 5.5471, "step": 34000 }, { "epoch": 9.38, "learning_rate": 4.916205858198741e-05, "loss": 5.4798, "step": 34500 }, { "epoch": 9.52, "learning_rate": 4.9144949356693135e-05, "loss": 5.2155, "step": 35000 }, { "epoch": 9.65, "learning_rate": 4.912784013139885e-05, "loss": 5.0368, "step": 35500 }, { "epoch": 9.79, "learning_rate": 4.911073090610458e-05, "loss": 4.8597, "step": 36000 }, { "epoch": 9.92, "learning_rate": 4.909365589926088e-05, "loss": 4.6933, "step": 36500 }, { "epoch": 10.06, "learning_rate": 4.907654667396661e-05, "loss": 4.5374, "step": 37000 }, { "epoch": 10.2, "learning_rate": 4.905943744867232e-05, "loss": 4.3855, "step": 37500 }, { "epoch": 10.33, "learning_rate": 4.904232822337805e-05, "loss": 4.2293, "step": 38000 }, { "epoch": 10.47, "learning_rate": 4.9025253216534355e-05, "loss": 4.0771, "step": 38500 }, { "epoch": 10.6, "learning_rate": 4.9008143991240076e-05, "loss": 3.9247, "step": 39000 }, { "epoch": 10.74, "learning_rate": 4.89910347659458e-05, "loss": 3.7601, "step": 39500 }, { "epoch": 10.88, "learning_rate": 4.897392554065152e-05, "loss": 3.3989, "step": 40000 }, { "epoch": 11.01, "learning_rate": 4.895685053380783e-05, "loss": 2.6924, "step": 40500 }, { "epoch": 11.15, "learning_rate": 4.893974130851355e-05, "loss": 2.3507, "step": 41000 }, { "epoch": 11.28, "learning_rate": 4.892263208321927e-05, "loss": 2.1874, "step": 41500 }, { "epoch": 11.42, "learning_rate": 4.890555707637558e-05, "loss": 2.0827, "step": 42000 }, { "epoch": 11.56, "learning_rate": 4.8888447851081304e-05, "loss": 2.001, "step": 42500 }, { "epoch": 11.69, "learning_rate": 4.8871338625787025e-05, "loss": 1.9375, "step": 43000 }, { "epoch": 11.83, "learning_rate": 4.8854229400492746e-05, "loss": 1.8864, "step": 43500 }, { "epoch": 11.96, "learning_rate": 4.8837154393649057e-05, "loss": 1.8373, "step": 44000 }, { "epoch": 12.1, "learning_rate": 4.882004516835478e-05, "loss": 1.7978, "step": 44500 }, { "epoch": 12.23, "learning_rate": 4.88029359430605e-05, "loss": 1.7603, "step": 45000 }, { "epoch": 12.37, "learning_rate": 4.878582671776622e-05, "loss": 1.7258, "step": 45500 }, { "epoch": 12.51, "learning_rate": 4.876875171092253e-05, "loss": 1.6949, "step": 46000 }, { "epoch": 12.64, "learning_rate": 4.875167670407884e-05, "loss": 1.6671, "step": 46500 }, { "epoch": 12.78, "learning_rate": 4.873456747878456e-05, "loss": 1.6404, "step": 47000 }, { "epoch": 12.91, "learning_rate": 4.8717458253490284e-05, "loss": 1.6187, "step": 47500 }, { "epoch": 13.05, "learning_rate": 4.8700349028196005e-05, "loss": 1.5947, "step": 48000 }, { "epoch": 13.19, "learning_rate": 4.8683239802901726e-05, "loss": 1.5738, "step": 48500 }, { "epoch": 13.32, "learning_rate": 4.866613057760745e-05, "loss": 1.5511, "step": 49000 }, { "epoch": 13.46, "learning_rate": 4.864902135231317e-05, "loss": 1.5372, "step": 49500 }, { "epoch": 13.59, "learning_rate": 4.863191212701889e-05, "loss": 1.5169, "step": 50000 }, { "epoch": 13.73, "learning_rate": 4.86148371201752e-05, "loss": 1.5009, "step": 50500 }, { "epoch": 13.87, "learning_rate": 4.859772789488092e-05, "loss": 1.4868, "step": 51000 }, { "epoch": 14.0, "learning_rate": 4.858061866958664e-05, "loss": 1.4705, "step": 51500 }, { "epoch": 14.14, "learning_rate": 4.8563509444292364e-05, "loss": 1.4536, "step": 52000 }, { "epoch": 14.27, "learning_rate": 4.8546400218998085e-05, "loss": 1.4392, "step": 52500 }, { "epoch": 14.41, "learning_rate": 4.8529359430604985e-05, "loss": 1.4263, "step": 53000 }, { "epoch": 14.55, "learning_rate": 4.8512250205310706e-05, "loss": 1.4116, "step": 53500 }, { "epoch": 14.68, "learning_rate": 4.849514098001643e-05, "loss": 1.3983, "step": 54000 }, { "epoch": 14.82, "learning_rate": 4.847803175472215e-05, "loss": 1.3863, "step": 54500 }, { "epoch": 14.95, "learning_rate": 4.846092252942787e-05, "loss": 1.3787, "step": 55000 }, { "epoch": 15.09, "learning_rate": 4.844381330413359e-05, "loss": 1.3675, "step": 55500 }, { "epoch": 15.23, "learning_rate": 4.84267382972899e-05, "loss": 1.3571, "step": 56000 }, { "epoch": 15.36, "learning_rate": 4.840962907199562e-05, "loss": 1.3476, "step": 56500 }, { "epoch": 15.5, "learning_rate": 4.8392519846701344e-05, "loss": 1.3398, "step": 57000 }, { "epoch": 15.63, "learning_rate": 4.8375410621407065e-05, "loss": 1.3287, "step": 57500 }, { "epoch": 15.77, "learning_rate": 4.8358301396112786e-05, "loss": 1.3205, "step": 58000 }, { "epoch": 15.91, "learning_rate": 4.834119217081851e-05, "loss": 1.3132, "step": 58500 }, { "epoch": 16.04, "learning_rate": 4.832408294552423e-05, "loss": 1.3066, "step": 59000 }, { "epoch": 16.18, "learning_rate": 4.830697372022995e-05, "loss": 1.2967, "step": 59500 }, { "epoch": 16.31, "learning_rate": 4.828986449493567e-05, "loss": 1.2888, "step": 60000 }, { "epoch": 16.45, "learning_rate": 4.827278948809198e-05, "loss": 1.284, "step": 60500 }, { "epoch": 16.59, "learning_rate": 4.82556802627977e-05, "loss": 1.2765, "step": 61000 }, { "epoch": 16.72, "learning_rate": 4.8238571037503424e-05, "loss": 1.2695, "step": 61500 }, { "epoch": 16.86, "learning_rate": 4.8221461812209145e-05, "loss": 1.2617, "step": 62000 }, { "epoch": 16.99, "learning_rate": 4.8204386805365456e-05, "loss": 1.2541, "step": 62500 }, { "epoch": 17.13, "learning_rate": 4.818727758007118e-05, "loss": 1.2493, "step": 63000 }, { "epoch": 17.26, "learning_rate": 4.81701683547769e-05, "loss": 1.2412, "step": 63500 }, { "epoch": 17.4, "learning_rate": 4.815305912948262e-05, "loss": 1.2379, "step": 64000 }, { "epoch": 17.54, "learning_rate": 4.813594990418834e-05, "loss": 1.2318, "step": 64500 }, { "epoch": 17.67, "learning_rate": 4.811887489734465e-05, "loss": 1.2239, "step": 65000 }, { "epoch": 17.81, "learning_rate": 4.810176567205037e-05, "loss": 1.2199, "step": 65500 }, { "epoch": 17.94, "learning_rate": 4.808465644675609e-05, "loss": 1.2146, "step": 66000 }, { "epoch": 18.08, "learning_rate": 4.8067581439912404e-05, "loss": 1.2083, "step": 66500 }, { "epoch": 18.22, "learning_rate": 4.8050472214618125e-05, "loss": 1.2046, "step": 67000 }, { "epoch": 18.35, "learning_rate": 4.8033362989323846e-05, "loss": 1.1977, "step": 67500 }, { "epoch": 18.49, "learning_rate": 4.801625376402957e-05, "loss": 1.1926, "step": 68000 }, { "epoch": 18.62, "learning_rate": 4.799914453873529e-05, "loss": 1.1862, "step": 68500 }, { "epoch": 18.76, "learning_rate": 4.798203531344101e-05, "loss": 1.1821, "step": 69000 }, { "epoch": 18.9, "learning_rate": 4.796496030659732e-05, "loss": 1.1815, "step": 69500 }, { "epoch": 19.03, "learning_rate": 4.794785108130304e-05, "loss": 1.1741, "step": 70000 }, { "epoch": 19.17, "learning_rate": 4.793074185600876e-05, "loss": 1.1687, "step": 70500 }, { "epoch": 19.3, "learning_rate": 4.7913632630714484e-05, "loss": 1.166, "step": 71000 }, { "epoch": 19.44, "learning_rate": 4.7896557623870795e-05, "loss": 1.1619, "step": 71500 }, { "epoch": 19.58, "learning_rate": 4.7879448398576516e-05, "loss": 1.1554, "step": 72000 }, { "epoch": 19.71, "learning_rate": 4.786233917328223e-05, "loss": 1.1516, "step": 72500 }, { "epoch": 19.85, "learning_rate": 4.784522994798796e-05, "loss": 1.1487, "step": 73000 }, { "epoch": 19.98, "learning_rate": 4.782815494114426e-05, "loss": 1.1398, "step": 73500 }, { "epoch": 20.12, "learning_rate": 4.781104571584999e-05, "loss": 1.1383, "step": 74000 }, { "epoch": 20.26, "learning_rate": 4.7793936490555704e-05, "loss": 1.1351, "step": 74500 }, { "epoch": 20.39, "learning_rate": 4.777682726526143e-05, "loss": 1.1313, "step": 75000 }, { "epoch": 20.53, "learning_rate": 4.7759752258417736e-05, "loss": 1.1274, "step": 75500 }, { "epoch": 20.66, "learning_rate": 4.7742643033123464e-05, "loss": 1.1229, "step": 76000 }, { "epoch": 20.8, "learning_rate": 4.772553380782918e-05, "loss": 1.1199, "step": 76500 }, { "epoch": 20.94, "learning_rate": 4.7708424582534906e-05, "loss": 1.1179, "step": 77000 }, { "epoch": 21.07, "learning_rate": 4.769134957569121e-05, "loss": 1.1129, "step": 77500 }, { "epoch": 21.21, "learning_rate": 4.767424035039694e-05, "loss": 1.1104, "step": 78000 }, { "epoch": 21.34, "learning_rate": 4.765713112510265e-05, "loss": 1.1044, "step": 78500 }, { "epoch": 21.48, "learning_rate": 4.764002189980838e-05, "loss": 1.1028, "step": 79000 }, { "epoch": 21.62, "learning_rate": 4.76229126745141e-05, "loss": 1.0974, "step": 79500 }, { "epoch": 21.75, "learning_rate": 4.760580344921982e-05, "loss": 1.0953, "step": 80000 }, { "epoch": 21.89, "learning_rate": 4.7588694223925544e-05, "loss": 1.0926, "step": 80500 }, { "epoch": 22.02, "learning_rate": 4.7571584998631265e-05, "loss": 1.0895, "step": 81000 }, { "epoch": 22.16, "learning_rate": 4.7554475773336986e-05, "loss": 1.0855, "step": 81500 }, { "epoch": 22.29, "learning_rate": 4.75374007664933e-05, "loss": 1.0816, "step": 82000 }, { "epoch": 22.43, "learning_rate": 4.752029154119902e-05, "loss": 1.08, "step": 82500 }, { "epoch": 22.57, "learning_rate": 4.750318231590474e-05, "loss": 1.0775, "step": 83000 }, { "epoch": 22.7, "learning_rate": 4.748610730906104e-05, "loss": 1.0736, "step": 83500 }, { "epoch": 22.84, "learning_rate": 4.746899808376677e-05, "loss": 1.0715, "step": 84000 }, { "epoch": 22.97, "learning_rate": 4.745188885847249e-05, "loss": 1.0663, "step": 84500 }, { "epoch": 23.11, "learning_rate": 4.7434779633178214e-05, "loss": 1.0654, "step": 85000 }, { "epoch": 23.25, "learning_rate": 4.741770462633452e-05, "loss": 1.0603, "step": 85500 }, { "epoch": 23.38, "learning_rate": 4.7400595401040245e-05, "loss": 1.061, "step": 86000 }, { "epoch": 23.52, "learning_rate": 4.7383486175745967e-05, "loss": 1.0583, "step": 86500 }, { "epoch": 23.65, "learning_rate": 4.736637695045169e-05, "loss": 1.0546, "step": 87000 }, { "epoch": 23.79, "learning_rate": 4.734926772515741e-05, "loss": 1.0525, "step": 87500 }, { "epoch": 23.93, "learning_rate": 4.733215849986313e-05, "loss": 1.0493, "step": 88000 }, { "epoch": 24.06, "learning_rate": 4.731508349301944e-05, "loss": 1.0467, "step": 88500 }, { "epoch": 24.2, "learning_rate": 4.729797426772516e-05, "loss": 1.0433, "step": 89000 }, { "epoch": 24.33, "learning_rate": 4.728086504243088e-05, "loss": 1.0436, "step": 89500 }, { "epoch": 24.47, "learning_rate": 4.7263790035587194e-05, "loss": 1.0389, "step": 90000 }, { "epoch": 24.61, "learning_rate": 4.7246680810292915e-05, "loss": 1.0378, "step": 90500 }, { "epoch": 24.74, "learning_rate": 4.7229571584998636e-05, "loss": 1.0358, "step": 91000 }, { "epoch": 24.88, "learning_rate": 4.721246235970436e-05, "loss": 1.0338, "step": 91500 }, { "epoch": 25.01, "learning_rate": 4.719535313441008e-05, "loss": 1.0303, "step": 92000 }, { "epoch": 25.15, "learning_rate": 4.71782439091158e-05, "loss": 1.0277, "step": 92500 }, { "epoch": 25.29, "learning_rate": 4.716113468382152e-05, "loss": 1.0288, "step": 93000 }, { "epoch": 25.42, "learning_rate": 4.714402545852724e-05, "loss": 1.0229, "step": 93500 }, { "epoch": 25.56, "learning_rate": 4.7126950451683546e-05, "loss": 1.0254, "step": 94000 }, { "epoch": 25.69, "learning_rate": 4.7109841226389274e-05, "loss": 1.0194, "step": 94500 }, { "epoch": 25.83, "learning_rate": 4.709273200109499e-05, "loss": 1.0182, "step": 95000 }, { "epoch": 25.97, "learning_rate": 4.7075622775800716e-05, "loss": 1.0175, "step": 95500 } ], "max_steps": 1471200, "num_train_epochs": 400, "total_flos": 2.5771642200931697e+19, "trial_name": null, "trial_params": null }