{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 21.05263157894737, "eval_steps": 500, "global_step": 6000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.08771929824561403, "grad_norm": 14.743780136108398, "learning_rate": 1.4942583732057417e-05, "loss": 0.9366, "step": 25 }, { "epoch": 0.17543859649122806, "grad_norm": 22.20017433166504, "learning_rate": 1.4882775119617225e-05, "loss": 1.2015, "step": 50 }, { "epoch": 0.2631578947368421, "grad_norm": 25.090787887573242, "learning_rate": 1.4822966507177032e-05, "loss": 1.0156, "step": 75 }, { "epoch": 0.3508771929824561, "grad_norm": 13.580609321594238, "learning_rate": 1.4763157894736842e-05, "loss": 1.117, "step": 100 }, { "epoch": 0.43859649122807015, "grad_norm": 12.062329292297363, "learning_rate": 1.470334928229665e-05, "loss": 0.8312, "step": 125 }, { "epoch": 0.5263157894736842, "grad_norm": 18.625316619873047, "learning_rate": 1.464354066985646e-05, "loss": 0.9938, "step": 150 }, { "epoch": 0.6140350877192983, "grad_norm": 11.628118515014648, "learning_rate": 1.4583732057416269e-05, "loss": 0.8922, "step": 175 }, { "epoch": 0.7017543859649122, "grad_norm": 12.121639251708984, "learning_rate": 1.4523923444976078e-05, "loss": 0.9787, "step": 200 }, { "epoch": 0.7894736842105263, "grad_norm": 14.091375350952148, "learning_rate": 1.4464114832535886e-05, "loss": 0.806, "step": 225 }, { "epoch": 0.8771929824561403, "grad_norm": 9.919096946716309, "learning_rate": 1.4404306220095696e-05, "loss": 0.9935, "step": 250 }, { "epoch": 0.9649122807017544, "grad_norm": 13.80988597869873, "learning_rate": 1.4344497607655503e-05, "loss": 0.8116, "step": 275 }, { "epoch": 1.0526315789473684, "grad_norm": 13.865740776062012, "learning_rate": 1.4284688995215311e-05, "loss": 0.7288, "step": 300 }, { "epoch": 1.1403508771929824, "grad_norm": 9.76303768157959, "learning_rate": 1.4224880382775121e-05, "loss": 0.8283, "step": 325 }, { "epoch": 1.2280701754385965, "grad_norm": 9.364381790161133, "learning_rate": 1.4165071770334929e-05, "loss": 0.6827, "step": 350 }, { "epoch": 1.3157894736842106, "grad_norm": 13.79258918762207, "learning_rate": 1.4105263157894738e-05, "loss": 0.7745, "step": 375 }, { "epoch": 1.4035087719298245, "grad_norm": 9.642690658569336, "learning_rate": 1.4045454545454546e-05, "loss": 0.7514, "step": 400 }, { "epoch": 1.4912280701754386, "grad_norm": 14.490239143371582, "learning_rate": 1.3985645933014356e-05, "loss": 0.6594, "step": 425 }, { "epoch": 1.5789473684210527, "grad_norm": 13.364012718200684, "learning_rate": 1.3925837320574163e-05, "loss": 0.6036, "step": 450 }, { "epoch": 1.6666666666666665, "grad_norm": 6.50977087020874, "learning_rate": 1.3866028708133971e-05, "loss": 0.7616, "step": 475 }, { "epoch": 1.7543859649122808, "grad_norm": 30.46417808532715, "learning_rate": 1.380622009569378e-05, "loss": 0.6735, "step": 500 }, { "epoch": 1.8421052631578947, "grad_norm": 17.846593856811523, "learning_rate": 1.3746411483253589e-05, "loss": 0.5902, "step": 525 }, { "epoch": 1.9298245614035088, "grad_norm": 19.83995819091797, "learning_rate": 1.3686602870813398e-05, "loss": 0.5928, "step": 550 }, { "epoch": 2.017543859649123, "grad_norm": 6.053647994995117, "learning_rate": 1.3626794258373206e-05, "loss": 0.5816, "step": 575 }, { "epoch": 2.1052631578947367, "grad_norm": 9.594230651855469, "learning_rate": 1.3566985645933015e-05, "loss": 0.488, "step": 600 }, { "epoch": 2.192982456140351, "grad_norm": 13.84262752532959, "learning_rate": 1.3507177033492823e-05, "loss": 0.568, "step": 625 }, { "epoch": 2.280701754385965, "grad_norm": 10.86662483215332, "learning_rate": 1.3447368421052633e-05, "loss": 0.4017, "step": 650 }, { "epoch": 2.3684210526315788, "grad_norm": 16.06069564819336, "learning_rate": 1.338755980861244e-05, "loss": 0.5112, "step": 675 }, { "epoch": 2.456140350877193, "grad_norm": 6.421328544616699, "learning_rate": 1.3327751196172248e-05, "loss": 0.3609, "step": 700 }, { "epoch": 2.543859649122807, "grad_norm": 6.225348949432373, "learning_rate": 1.3267942583732058e-05, "loss": 0.5531, "step": 725 }, { "epoch": 2.6315789473684212, "grad_norm": 14.942207336425781, "learning_rate": 1.3208133971291866e-05, "loss": 0.4272, "step": 750 }, { "epoch": 2.719298245614035, "grad_norm": 8.975879669189453, "learning_rate": 1.3148325358851675e-05, "loss": 0.4509, "step": 775 }, { "epoch": 2.807017543859649, "grad_norm": 21.482088088989258, "learning_rate": 1.3088516746411483e-05, "loss": 0.4233, "step": 800 }, { "epoch": 2.8947368421052633, "grad_norm": 5.1405792236328125, "learning_rate": 1.3028708133971293e-05, "loss": 0.418, "step": 825 }, { "epoch": 2.982456140350877, "grad_norm": 14.574483871459961, "learning_rate": 1.29688995215311e-05, "loss": 0.6778, "step": 850 }, { "epoch": 3.0701754385964914, "grad_norm": 7.385740280151367, "learning_rate": 1.290909090909091e-05, "loss": 0.4127, "step": 875 }, { "epoch": 3.1578947368421053, "grad_norm": 48.35734939575195, "learning_rate": 1.2849282296650718e-05, "loss": 0.4752, "step": 900 }, { "epoch": 3.245614035087719, "grad_norm": 8.056925773620605, "learning_rate": 1.2789473684210526e-05, "loss": 0.411, "step": 925 }, { "epoch": 3.3333333333333335, "grad_norm": 20.510982513427734, "learning_rate": 1.2729665071770335e-05, "loss": 0.495, "step": 950 }, { "epoch": 3.4210526315789473, "grad_norm": 25.063684463500977, "learning_rate": 1.2669856459330143e-05, "loss": 0.2436, "step": 975 }, { "epoch": 3.5087719298245617, "grad_norm": 15.523259162902832, "learning_rate": 1.2610047846889953e-05, "loss": 0.4612, "step": 1000 }, { "epoch": 3.5964912280701755, "grad_norm": 1.9646857976913452, "learning_rate": 1.255023923444976e-05, "loss": 0.4426, "step": 1025 }, { "epoch": 3.6842105263157894, "grad_norm": 30.676250457763672, "learning_rate": 1.249043062200957e-05, "loss": 0.3333, "step": 1050 }, { "epoch": 3.7719298245614032, "grad_norm": 6.475268363952637, "learning_rate": 1.2430622009569378e-05, "loss": 0.2945, "step": 1075 }, { "epoch": 3.8596491228070176, "grad_norm": 24.684438705444336, "learning_rate": 1.2370813397129186e-05, "loss": 0.4339, "step": 1100 }, { "epoch": 3.9473684210526314, "grad_norm": 38.601226806640625, "learning_rate": 1.2311004784688995e-05, "loss": 0.3701, "step": 1125 }, { "epoch": 4.035087719298246, "grad_norm": 2.8431146144866943, "learning_rate": 1.2251196172248803e-05, "loss": 0.2782, "step": 1150 }, { "epoch": 4.12280701754386, "grad_norm": 23.384532928466797, "learning_rate": 1.2191387559808613e-05, "loss": 0.312, "step": 1175 }, { "epoch": 4.2105263157894735, "grad_norm": 3.222729444503784, "learning_rate": 1.213157894736842e-05, "loss": 0.264, "step": 1200 }, { "epoch": 4.298245614035087, "grad_norm": 3.5861947536468506, "learning_rate": 1.207177033492823e-05, "loss": 0.3104, "step": 1225 }, { "epoch": 4.385964912280702, "grad_norm": 4.091048717498779, "learning_rate": 1.2011961722488038e-05, "loss": 0.3567, "step": 1250 }, { "epoch": 4.473684210526316, "grad_norm": 1.3814598321914673, "learning_rate": 1.1952153110047847e-05, "loss": 0.3455, "step": 1275 }, { "epoch": 4.56140350877193, "grad_norm": 2.480316162109375, "learning_rate": 1.1892344497607655e-05, "loss": 0.1769, "step": 1300 }, { "epoch": 4.649122807017544, "grad_norm": 20.069162368774414, "learning_rate": 1.1832535885167463e-05, "loss": 0.2483, "step": 1325 }, { "epoch": 4.7368421052631575, "grad_norm": 1.7885949611663818, "learning_rate": 1.1772727272727272e-05, "loss": 0.2283, "step": 1350 }, { "epoch": 4.824561403508772, "grad_norm": 40.37774658203125, "learning_rate": 1.171291866028708e-05, "loss": 0.2277, "step": 1375 }, { "epoch": 4.912280701754386, "grad_norm": 2.960739850997925, "learning_rate": 1.1653110047846891e-05, "loss": 0.2077, "step": 1400 }, { "epoch": 5.0, "grad_norm": 14.10015869140625, "learning_rate": 1.15933014354067e-05, "loss": 0.3859, "step": 1425 }, { "epoch": 5.087719298245614, "grad_norm": 4.440461158752441, "learning_rate": 1.1533492822966509e-05, "loss": 0.1354, "step": 1450 }, { "epoch": 5.175438596491228, "grad_norm": 17.193849563598633, "learning_rate": 1.1473684210526317e-05, "loss": 0.2935, "step": 1475 }, { "epoch": 5.2631578947368425, "grad_norm": 1.2828125953674316, "learning_rate": 1.1413875598086125e-05, "loss": 0.1394, "step": 1500 }, { "epoch": 5.350877192982456, "grad_norm": 4.809077262878418, "learning_rate": 1.1354066985645934e-05, "loss": 0.3651, "step": 1525 }, { "epoch": 5.43859649122807, "grad_norm": 43.53572463989258, "learning_rate": 1.1294258373205742e-05, "loss": 0.1187, "step": 1550 }, { "epoch": 5.526315789473684, "grad_norm": 5.647329330444336, "learning_rate": 1.1234449760765551e-05, "loss": 0.2719, "step": 1575 }, { "epoch": 5.614035087719298, "grad_norm": 25.37187957763672, "learning_rate": 1.117464114832536e-05, "loss": 0.2477, "step": 1600 }, { "epoch": 5.701754385964913, "grad_norm": 4.228896617889404, "learning_rate": 1.1114832535885169e-05, "loss": 0.3608, "step": 1625 }, { "epoch": 5.7894736842105265, "grad_norm": 2.1760356426239014, "learning_rate": 1.1055023923444977e-05, "loss": 0.1223, "step": 1650 }, { "epoch": 5.87719298245614, "grad_norm": 28.078975677490234, "learning_rate": 1.0995215311004786e-05, "loss": 0.3207, "step": 1675 }, { "epoch": 5.964912280701754, "grad_norm": 12.583624839782715, "learning_rate": 1.0935406698564594e-05, "loss": 0.1094, "step": 1700 }, { "epoch": 6.052631578947368, "grad_norm": 0.4104742109775543, "learning_rate": 1.0875598086124402e-05, "loss": 0.1937, "step": 1725 }, { "epoch": 6.140350877192983, "grad_norm": 1.5135095119476318, "learning_rate": 1.0815789473684211e-05, "loss": 0.3113, "step": 1750 }, { "epoch": 6.228070175438597, "grad_norm": 1.9227218627929688, "learning_rate": 1.0755980861244019e-05, "loss": 0.1179, "step": 1775 }, { "epoch": 6.315789473684211, "grad_norm": 7.322267055511475, "learning_rate": 1.0696172248803829e-05, "loss": 0.209, "step": 1800 }, { "epoch": 6.4035087719298245, "grad_norm": 2.8994693756103516, "learning_rate": 1.0636363636363636e-05, "loss": 0.1743, "step": 1825 }, { "epoch": 6.491228070175438, "grad_norm": 21.65730857849121, "learning_rate": 1.0576555023923446e-05, "loss": 0.2483, "step": 1850 }, { "epoch": 6.578947368421053, "grad_norm": 24.210969924926758, "learning_rate": 1.0516746411483254e-05, "loss": 0.1348, "step": 1875 }, { "epoch": 6.666666666666667, "grad_norm": 1.589753270149231, "learning_rate": 1.0456937799043063e-05, "loss": 0.229, "step": 1900 }, { "epoch": 6.754385964912281, "grad_norm": 3.508387327194214, "learning_rate": 1.0397129186602871e-05, "loss": 0.0909, "step": 1925 }, { "epoch": 6.842105263157895, "grad_norm": 1.89371919631958, "learning_rate": 1.0337320574162679e-05, "loss": 0.1859, "step": 1950 }, { "epoch": 6.9298245614035086, "grad_norm": 0.35574591159820557, "learning_rate": 1.0277511961722489e-05, "loss": 0.1675, "step": 1975 }, { "epoch": 7.017543859649122, "grad_norm": 0.621152937412262, "learning_rate": 1.0217703349282296e-05, "loss": 0.1082, "step": 2000 }, { "epoch": 7.105263157894737, "grad_norm": 0.5694774985313416, "learning_rate": 1.0157894736842106e-05, "loss": 0.0886, "step": 2025 }, { "epoch": 7.192982456140351, "grad_norm": 7.276403427124023, "learning_rate": 1.0098086124401914e-05, "loss": 0.2697, "step": 2050 }, { "epoch": 7.280701754385965, "grad_norm": 0.4015137851238251, "learning_rate": 1.0038277511961723e-05, "loss": 0.1301, "step": 2075 }, { "epoch": 7.368421052631579, "grad_norm": 0.27876588702201843, "learning_rate": 9.978468899521531e-06, "loss": 0.0687, "step": 2100 }, { "epoch": 7.456140350877193, "grad_norm": 7.235930442810059, "learning_rate": 9.918660287081339e-06, "loss": 0.0734, "step": 2125 }, { "epoch": 7.543859649122807, "grad_norm": 1.6751177310943604, "learning_rate": 9.858851674641148e-06, "loss": 0.1574, "step": 2150 }, { "epoch": 7.631578947368421, "grad_norm": 0.8539127111434937, "learning_rate": 9.799043062200956e-06, "loss": 0.0483, "step": 2175 }, { "epoch": 7.719298245614035, "grad_norm": 0.26878660917282104, "learning_rate": 9.739234449760766e-06, "loss": 0.1632, "step": 2200 }, { "epoch": 7.807017543859649, "grad_norm": 106.50234985351562, "learning_rate": 9.679425837320574e-06, "loss": 0.1156, "step": 2225 }, { "epoch": 7.894736842105263, "grad_norm": 2.002487897872925, "learning_rate": 9.619617224880383e-06, "loss": 0.2036, "step": 2250 }, { "epoch": 7.982456140350877, "grad_norm": 0.5302895307540894, "learning_rate": 9.559808612440191e-06, "loss": 0.1417, "step": 2275 }, { "epoch": 8.070175438596491, "grad_norm": 0.314165323972702, "learning_rate": 9.5e-06, "loss": 0.0673, "step": 2300 }, { "epoch": 8.157894736842104, "grad_norm": 0.5242946743965149, "learning_rate": 9.440191387559808e-06, "loss": 0.0689, "step": 2325 }, { "epoch": 8.24561403508772, "grad_norm": 48.71388244628906, "learning_rate": 9.380382775119616e-06, "loss": 0.1863, "step": 2350 }, { "epoch": 8.333333333333334, "grad_norm": 0.5670514106750488, "learning_rate": 9.320574162679426e-06, "loss": 0.0823, "step": 2375 }, { "epoch": 8.421052631578947, "grad_norm": 0.6025238633155823, "learning_rate": 9.260765550239234e-06, "loss": 0.1118, "step": 2400 }, { "epoch": 8.508771929824562, "grad_norm": 42.649696350097656, "learning_rate": 9.200956937799043e-06, "loss": 0.1333, "step": 2425 }, { "epoch": 8.596491228070175, "grad_norm": 0.4079756438732147, "learning_rate": 9.141148325358851e-06, "loss": 0.156, "step": 2450 }, { "epoch": 8.68421052631579, "grad_norm": 0.31388023495674133, "learning_rate": 9.08133971291866e-06, "loss": 0.2159, "step": 2475 }, { "epoch": 8.771929824561404, "grad_norm": 0.2017977237701416, "learning_rate": 9.021531100478468e-06, "loss": 0.0936, "step": 2500 }, { "epoch": 8.859649122807017, "grad_norm": 2.2358014583587646, "learning_rate": 6.697368421052631e-06, "loss": 0.1884, "step": 2525 }, { "epoch": 8.947368421052632, "grad_norm": 0.5669077038764954, "learning_rate": 6.615131578947369e-06, "loss": 0.2183, "step": 2550 }, { "epoch": 9.035087719298245, "grad_norm": 27.18621253967285, "learning_rate": 6.532894736842106e-06, "loss": 0.1232, "step": 2575 }, { "epoch": 9.12280701754386, "grad_norm": 468.5591735839844, "learning_rate": 6.450657894736842e-06, "loss": 0.0516, "step": 2600 }, { "epoch": 9.210526315789474, "grad_norm": 0.24007272720336914, "learning_rate": 6.368421052631579e-06, "loss": 0.0695, "step": 2625 }, { "epoch": 9.298245614035087, "grad_norm": 99.38546752929688, "learning_rate": 6.286184210526316e-06, "loss": 0.0947, "step": 2650 }, { "epoch": 9.385964912280702, "grad_norm": 3.8673267364501953, "learning_rate": 6.203947368421053e-06, "loss": 0.1062, "step": 2675 }, { "epoch": 9.473684210526315, "grad_norm": 0.2559914290904999, "learning_rate": 6.12171052631579e-06, "loss": 0.0523, "step": 2700 }, { "epoch": 9.56140350877193, "grad_norm": 0.12371934205293655, "learning_rate": 6.039473684210526e-06, "loss": 0.0495, "step": 2725 }, { "epoch": 9.649122807017545, "grad_norm": 2.5249898433685303, "learning_rate": 5.957236842105263e-06, "loss": 0.1678, "step": 2750 }, { "epoch": 9.736842105263158, "grad_norm": 0.2619442939758301, "learning_rate": 5.8750000000000005e-06, "loss": 0.0534, "step": 2775 }, { "epoch": 9.824561403508772, "grad_norm": 1.3705483675003052, "learning_rate": 5.792763157894737e-06, "loss": 0.1326, "step": 2800 }, { "epoch": 9.912280701754385, "grad_norm": 74.48184967041016, "learning_rate": 5.710526315789474e-06, "loss": 0.1795, "step": 2825 }, { "epoch": 10.0, "grad_norm": 79.78992462158203, "learning_rate": 5.6282894736842106e-06, "loss": 0.092, "step": 2850 }, { "epoch": 10.087719298245615, "grad_norm": 0.5286645889282227, "learning_rate": 5.546052631578947e-06, "loss": 0.0418, "step": 2875 }, { "epoch": 10.175438596491228, "grad_norm": 109.82157135009766, "learning_rate": 5.463815789473684e-06, "loss": 0.193, "step": 2900 }, { "epoch": 10.263157894736842, "grad_norm": 3.963012933731079, "learning_rate": 5.3815789473684215e-06, "loss": 0.0897, "step": 2925 }, { "epoch": 10.350877192982455, "grad_norm": 0.41540616750717163, "learning_rate": 5.299342105263158e-06, "loss": 0.1723, "step": 2950 }, { "epoch": 10.43859649122807, "grad_norm": 5.900774002075195, "learning_rate": 5.217105263157895e-06, "loss": 0.0784, "step": 2975 }, { "epoch": 10.526315789473685, "grad_norm": 12.152785301208496, "learning_rate": 5.1348684210526315e-06, "loss": 0.1186, "step": 3000 }, { "epoch": 10.614035087719298, "grad_norm": 0.21494676172733307, "learning_rate": 3.6315789473684213e-06, "loss": 0.0358, "step": 3025 }, { "epoch": 10.701754385964913, "grad_norm": 90.48876953125, "learning_rate": 3.537593984962406e-06, "loss": 0.0471, "step": 3050 }, { "epoch": 10.789473684210526, "grad_norm": 0.23261310160160065, "learning_rate": 3.4436090225563912e-06, "loss": 0.0276, "step": 3075 }, { "epoch": 10.87719298245614, "grad_norm": 0.782564640045166, "learning_rate": 3.349624060150376e-06, "loss": 0.0718, "step": 3100 }, { "epoch": 10.964912280701755, "grad_norm": 0.33198878169059753, "learning_rate": 3.255639097744361e-06, "loss": 0.0564, "step": 3125 }, { "epoch": 11.052631578947368, "grad_norm": 0.14864841103553772, "learning_rate": 3.161654135338346e-06, "loss": 0.0508, "step": 3150 }, { "epoch": 11.140350877192983, "grad_norm": 0.1588129848241806, "learning_rate": 3.067669172932331e-06, "loss": 0.1641, "step": 3175 }, { "epoch": 11.228070175438596, "grad_norm": 0.6562497019767761, "learning_rate": 2.973684210526316e-06, "loss": 0.0544, "step": 3200 }, { "epoch": 11.31578947368421, "grad_norm": 0.3608649969100952, "learning_rate": 2.879699248120301e-06, "loss": 0.0391, "step": 3225 }, { "epoch": 11.403508771929825, "grad_norm": 20.602819442749023, "learning_rate": 2.785714285714286e-06, "loss": 0.0338, "step": 3250 }, { "epoch": 11.491228070175438, "grad_norm": 157.23171997070312, "learning_rate": 2.691729323308271e-06, "loss": 0.0771, "step": 3275 }, { "epoch": 11.578947368421053, "grad_norm": 0.22338764369487762, "learning_rate": 2.597744360902256e-06, "loss": 0.1187, "step": 3300 }, { "epoch": 11.666666666666666, "grad_norm": 57.78382110595703, "learning_rate": 2.503759398496241e-06, "loss": 0.0165, "step": 3325 }, { "epoch": 11.75438596491228, "grad_norm": 7.400601863861084, "learning_rate": 2.4097744360902257e-06, "loss": 0.0092, "step": 3350 }, { "epoch": 11.842105263157894, "grad_norm": 0.18190287053585052, "learning_rate": 2.3157894736842105e-06, "loss": 0.1059, "step": 3375 }, { "epoch": 11.929824561403509, "grad_norm": 0.1269238144159317, "learning_rate": 2.2218045112781957e-06, "loss": 0.0571, "step": 3400 }, { "epoch": 12.017543859649123, "grad_norm": 0.10843072831630707, "learning_rate": 2.1278195488721805e-06, "loss": 0.1341, "step": 3425 }, { "epoch": 12.105263157894736, "grad_norm": 0.7199507355690002, "learning_rate": 2.0338345864661656e-06, "loss": 0.0063, "step": 3450 }, { "epoch": 12.192982456140351, "grad_norm": 0.19009971618652344, "learning_rate": 1.9398496240601504e-06, "loss": 0.054, "step": 3475 }, { "epoch": 12.280701754385966, "grad_norm": 0.18020951747894287, "learning_rate": 1.8458646616541354e-06, "loss": 0.0148, "step": 3500 }, { "epoch": 12.368421052631579, "grad_norm": 0.2494085282087326, "learning_rate": 6.569377990430622e-06, "loss": 0.1168, "step": 3525 }, { "epoch": 12.456140350877194, "grad_norm": 0.16571170091629028, "learning_rate": 6.509569377990431e-06, "loss": 0.0482, "step": 3550 }, { "epoch": 12.543859649122806, "grad_norm": 0.10973577946424484, "learning_rate": 6.44976076555024e-06, "loss": 0.1372, "step": 3575 }, { "epoch": 12.631578947368421, "grad_norm": 0.14920943975448608, "learning_rate": 6.389952153110048e-06, "loss": 0.0791, "step": 3600 }, { "epoch": 12.719298245614034, "grad_norm": 4.5087504386901855, "learning_rate": 6.330143540669857e-06, "loss": 0.1614, "step": 3625 }, { "epoch": 12.807017543859649, "grad_norm": 0.11903402209281921, "learning_rate": 6.270334928229666e-06, "loss": 0.0082, "step": 3650 }, { "epoch": 12.894736842105264, "grad_norm": 0.24349439144134521, "learning_rate": 6.210526315789474e-06, "loss": 0.1936, "step": 3675 }, { "epoch": 12.982456140350877, "grad_norm": 146.2913360595703, "learning_rate": 6.150717703349282e-06, "loss": 0.1465, "step": 3700 }, { "epoch": 13.070175438596491, "grad_norm": 88.66224670410156, "learning_rate": 6.090909090909091e-06, "loss": 0.0751, "step": 3725 }, { "epoch": 13.157894736842104, "grad_norm": 0.11558452993631363, "learning_rate": 6.0311004784689e-06, "loss": 0.0652, "step": 3750 }, { "epoch": 13.24561403508772, "grad_norm": 3.1845107078552246, "learning_rate": 5.971291866028708e-06, "loss": 0.0969, "step": 3775 }, { "epoch": 13.333333333333334, "grad_norm": 0.5500511527061462, "learning_rate": 5.911483253588517e-06, "loss": 0.107, "step": 3800 }, { "epoch": 13.421052631578947, "grad_norm": 0.08417954295873642, "learning_rate": 5.851674641148326e-06, "loss": 0.0165, "step": 3825 }, { "epoch": 13.508771929824562, "grad_norm": 0.0910855233669281, "learning_rate": 5.791866028708134e-06, "loss": 0.1326, "step": 3850 }, { "epoch": 13.596491228070175, "grad_norm": 65.31088256835938, "learning_rate": 5.732057416267942e-06, "loss": 0.0365, "step": 3875 }, { "epoch": 13.68421052631579, "grad_norm": 0.09439278393983841, "learning_rate": 5.672248803827751e-06, "loss": 0.0784, "step": 3900 }, { "epoch": 13.771929824561404, "grad_norm": 0.17843316495418549, "learning_rate": 5.6124401913875595e-06, "loss": 0.0232, "step": 3925 }, { "epoch": 13.859649122807017, "grad_norm": 0.16249841451644897, "learning_rate": 5.552631578947368e-06, "loss": 0.0107, "step": 3950 }, { "epoch": 13.947368421052632, "grad_norm": 0.545814573764801, "learning_rate": 5.492822966507177e-06, "loss": 0.0947, "step": 3975 }, { "epoch": 14.035087719298245, "grad_norm": 0.7138233184814453, "learning_rate": 5.433014354066986e-06, "loss": 0.1812, "step": 4000 }, { "epoch": 14.12280701754386, "grad_norm": 381.4528503417969, "learning_rate": 5.373205741626794e-06, "loss": 0.0399, "step": 4025 }, { "epoch": 14.210526315789474, "grad_norm": 0.053043171763420105, "learning_rate": 5.313397129186603e-06, "loss": 0.1249, "step": 4050 }, { "epoch": 14.298245614035087, "grad_norm": 1.4563120603561401, "learning_rate": 5.253588516746412e-06, "loss": 0.0361, "step": 4075 }, { "epoch": 14.385964912280702, "grad_norm": 0.06027642637491226, "learning_rate": 5.19377990430622e-06, "loss": 0.0799, "step": 4100 }, { "epoch": 14.473684210526315, "grad_norm": 0.11955183744430542, "learning_rate": 5.133971291866029e-06, "loss": 0.2006, "step": 4125 }, { "epoch": 14.56140350877193, "grad_norm": 0.07738160341978073, "learning_rate": 5.074162679425838e-06, "loss": 0.16, "step": 4150 }, { "epoch": 14.649122807017545, "grad_norm": 28.58780860900879, "learning_rate": 5.014354066985646e-06, "loss": 0.0675, "step": 4175 }, { "epoch": 14.736842105263158, "grad_norm": 0.08654920011758804, "learning_rate": 4.954545454545455e-06, "loss": 0.0584, "step": 4200 }, { "epoch": 14.824561403508772, "grad_norm": 0.1421671062707901, "learning_rate": 4.894736842105264e-06, "loss": 0.0049, "step": 4225 }, { "epoch": 14.912280701754385, "grad_norm": 0.23447291553020477, "learning_rate": 4.834928229665072e-06, "loss": 0.0122, "step": 4250 }, { "epoch": 15.0, "grad_norm": 54.224788665771484, "learning_rate": 4.775119617224881e-06, "loss": 0.0921, "step": 4275 }, { "epoch": 15.087719298245615, "grad_norm": 0.08427131175994873, "learning_rate": 4.715311004784689e-06, "loss": 0.0061, "step": 4300 }, { "epoch": 15.175438596491228, "grad_norm": 0.232700914144516, "learning_rate": 4.6555023923444975e-06, "loss": 0.1011, "step": 4325 }, { "epoch": 15.263157894736842, "grad_norm": 0.09075941890478134, "learning_rate": 4.595693779904306e-06, "loss": 0.0419, "step": 4350 }, { "epoch": 15.350877192982455, "grad_norm": 0.20714852213859558, "learning_rate": 4.535885167464115e-06, "loss": 0.1786, "step": 4375 }, { "epoch": 15.43859649122807, "grad_norm": 0.07586350291967392, "learning_rate": 4.476076555023924e-06, "loss": 0.0314, "step": 4400 }, { "epoch": 15.526315789473685, "grad_norm": 0.29287371039390564, "learning_rate": 4.416267942583732e-06, "loss": 0.0635, "step": 4425 }, { "epoch": 15.614035087719298, "grad_norm": 50.64339065551758, "learning_rate": 4.356459330143541e-06, "loss": 0.0854, "step": 4450 }, { "epoch": 15.701754385964913, "grad_norm": 0.12661752104759216, "learning_rate": 4.29665071770335e-06, "loss": 0.1378, "step": 4475 }, { "epoch": 15.789473684210526, "grad_norm": 2.418790340423584, "learning_rate": 4.2368421052631575e-06, "loss": 0.0325, "step": 4500 }, { "epoch": 15.87719298245614, "grad_norm": 231.30796813964844, "learning_rate": 4.177033492822966e-06, "loss": 0.1398, "step": 4525 }, { "epoch": 15.964912280701755, "grad_norm": 0.23073101043701172, "learning_rate": 4.117224880382775e-06, "loss": 0.0237, "step": 4550 }, { "epoch": 16.05263157894737, "grad_norm": 0.044060900807380676, "learning_rate": 4.0574162679425835e-06, "loss": 0.0074, "step": 4575 }, { "epoch": 16.140350877192983, "grad_norm": 1.4264813661575317, "learning_rate": 3.997607655502392e-06, "loss": 0.025, "step": 4600 }, { "epoch": 16.228070175438596, "grad_norm": 0.41092243790626526, "learning_rate": 3.937799043062201e-06, "loss": 0.0049, "step": 4625 }, { "epoch": 16.31578947368421, "grad_norm": 3.6420750617980957, "learning_rate": 3.8779904306220095e-06, "loss": 0.0491, "step": 4650 }, { "epoch": 16.403508771929825, "grad_norm": 0.08417033404111862, "learning_rate": 3.818181818181818e-06, "loss": 0.003, "step": 4675 }, { "epoch": 16.49122807017544, "grad_norm": 0.41752633452415466, "learning_rate": 3.7583732057416273e-06, "loss": 0.0649, "step": 4700 }, { "epoch": 16.57894736842105, "grad_norm": 0.04892294108867645, "learning_rate": 3.6985645933014356e-06, "loss": 0.1577, "step": 4725 }, { "epoch": 16.666666666666668, "grad_norm": 0.07886885106563568, "learning_rate": 3.6387559808612442e-06, "loss": 0.1133, "step": 4750 }, { "epoch": 16.75438596491228, "grad_norm": 202.08302307128906, "learning_rate": 3.5789473684210525e-06, "loss": 0.0151, "step": 4775 }, { "epoch": 16.842105263157894, "grad_norm": 77.71251678466797, "learning_rate": 3.519138755980861e-06, "loss": 0.0644, "step": 4800 }, { "epoch": 16.92982456140351, "grad_norm": 0.04525836929678917, "learning_rate": 3.45933014354067e-06, "loss": 0.0283, "step": 4825 }, { "epoch": 17.017543859649123, "grad_norm": 0.035384029150009155, "learning_rate": 3.399521531100479e-06, "loss": 0.0431, "step": 4850 }, { "epoch": 17.105263157894736, "grad_norm": 0.05048861354589462, "learning_rate": 3.339712918660287e-06, "loss": 0.0358, "step": 4875 }, { "epoch": 17.19298245614035, "grad_norm": 0.1347443163394928, "learning_rate": 3.279904306220096e-06, "loss": 0.0206, "step": 4900 }, { "epoch": 17.280701754385966, "grad_norm": 252.46458435058594, "learning_rate": 3.2200956937799046e-06, "loss": 0.0804, "step": 4925 }, { "epoch": 17.36842105263158, "grad_norm": 0.05461292341351509, "learning_rate": 3.1602870813397132e-06, "loss": 0.0784, "step": 4950 }, { "epoch": 17.45614035087719, "grad_norm": 2.0001866817474365, "learning_rate": 3.1004784688995215e-06, "loss": 0.0947, "step": 4975 }, { "epoch": 17.54385964912281, "grad_norm": 0.04590539261698723, "learning_rate": 3.04066985645933e-06, "loss": 0.075, "step": 5000 }, { "epoch": 17.63157894736842, "grad_norm": 0.04810748249292374, "learning_rate": 2.980861244019139e-06, "loss": 0.0342, "step": 5025 }, { "epoch": 17.719298245614034, "grad_norm": 0.8750467896461487, "learning_rate": 2.9210526315789475e-06, "loss": 0.0618, "step": 5050 }, { "epoch": 17.80701754385965, "grad_norm": 56.14554977416992, "learning_rate": 2.8612440191387558e-06, "loss": 0.026, "step": 5075 }, { "epoch": 17.894736842105264, "grad_norm": 0.8380083441734314, "learning_rate": 2.8014354066985645e-06, "loss": 0.0135, "step": 5100 }, { "epoch": 17.982456140350877, "grad_norm": 0.08012706786394119, "learning_rate": 2.741626794258373e-06, "loss": 0.0586, "step": 5125 }, { "epoch": 18.07017543859649, "grad_norm": 1.2549360990524292, "learning_rate": 2.6818181818181822e-06, "loss": 0.0047, "step": 5150 }, { "epoch": 18.157894736842106, "grad_norm": 0.03993632644414902, "learning_rate": 2.6220095693779905e-06, "loss": 0.0869, "step": 5175 }, { "epoch": 18.24561403508772, "grad_norm": 62.21980285644531, "learning_rate": 2.562200956937799e-06, "loss": 0.0987, "step": 5200 }, { "epoch": 18.333333333333332, "grad_norm": 0.10092537105083466, "learning_rate": 2.502392344497608e-06, "loss": 0.1554, "step": 5225 }, { "epoch": 18.42105263157895, "grad_norm": 0.7630422115325928, "learning_rate": 2.4425837320574165e-06, "loss": 0.0522, "step": 5250 }, { "epoch": 18.50877192982456, "grad_norm": 62.42081069946289, "learning_rate": 2.382775119617225e-06, "loss": 0.0298, "step": 5275 }, { "epoch": 18.596491228070175, "grad_norm": 0.046070296317338943, "learning_rate": 2.3229665071770335e-06, "loss": 0.1207, "step": 5300 }, { "epoch": 18.68421052631579, "grad_norm": 0.04533836990594864, "learning_rate": 2.263157894736842e-06, "loss": 0.146, "step": 5325 }, { "epoch": 18.771929824561404, "grad_norm": 0.03193026781082153, "learning_rate": 2.203349282296651e-06, "loss": 0.0106, "step": 5350 }, { "epoch": 18.859649122807017, "grad_norm": 0.03594537451863289, "learning_rate": 2.143540669856459e-06, "loss": 0.0266, "step": 5375 }, { "epoch": 18.94736842105263, "grad_norm": 0.07382319122552872, "learning_rate": 2.0837320574162678e-06, "loss": 0.0954, "step": 5400 }, { "epoch": 19.035087719298247, "grad_norm": 0.0792032852768898, "learning_rate": 2.0239234449760764e-06, "loss": 0.0871, "step": 5425 }, { "epoch": 19.12280701754386, "grad_norm": 0.06994396448135376, "learning_rate": 1.9641148325358855e-06, "loss": 0.0092, "step": 5450 }, { "epoch": 19.210526315789473, "grad_norm": 36.98837661743164, "learning_rate": 1.904306220095694e-06, "loss": 0.1442, "step": 5475 }, { "epoch": 19.29824561403509, "grad_norm": 0.11950813978910446, "learning_rate": 1.8444976076555023e-06, "loss": 0.0344, "step": 5500 }, { "epoch": 19.385964912280702, "grad_norm": 0.053549814969301224, "learning_rate": 1.7846889952153112e-06, "loss": 0.0609, "step": 5525 }, { "epoch": 19.473684210526315, "grad_norm": 0.05915123224258423, "learning_rate": 1.7248803827751196e-06, "loss": 0.0212, "step": 5550 }, { "epoch": 19.56140350877193, "grad_norm": 0.03509140759706497, "learning_rate": 1.6650717703349283e-06, "loss": 0.0019, "step": 5575 }, { "epoch": 19.649122807017545, "grad_norm": 0.04774395003914833, "learning_rate": 1.605263157894737e-06, "loss": 0.032, "step": 5600 }, { "epoch": 19.736842105263158, "grad_norm": 0.1909104436635971, "learning_rate": 1.5454545454545454e-06, "loss": 0.0351, "step": 5625 }, { "epoch": 19.82456140350877, "grad_norm": 0.11199568212032318, "learning_rate": 1.4856459330143541e-06, "loss": 0.0268, "step": 5650 }, { "epoch": 19.912280701754387, "grad_norm": 17.852022171020508, "learning_rate": 1.4258373205741628e-06, "loss": 0.0397, "step": 5675 }, { "epoch": 20.0, "grad_norm": 0.18822583556175232, "learning_rate": 1.3660287081339715e-06, "loss": 0.0541, "step": 5700 }, { "epoch": 20.087719298245613, "grad_norm": 50.338890075683594, "learning_rate": 1.30622009569378e-06, "loss": 0.016, "step": 5725 }, { "epoch": 20.17543859649123, "grad_norm": 0.05340347811579704, "learning_rate": 1.2464114832535886e-06, "loss": 0.0084, "step": 5750 }, { "epoch": 20.263157894736842, "grad_norm": 0.04629586637020111, "learning_rate": 1.186602870813397e-06, "loss": 0.0356, "step": 5775 }, { "epoch": 20.350877192982455, "grad_norm": 0.9073031544685364, "learning_rate": 1.1267942583732058e-06, "loss": 0.0786, "step": 5800 }, { "epoch": 20.43859649122807, "grad_norm": 10.095131874084473, "learning_rate": 1.0669856459330144e-06, "loss": 0.0419, "step": 5825 }, { "epoch": 20.526315789473685, "grad_norm": 0.16011740267276764, "learning_rate": 1.0071770334928231e-06, "loss": 0.0653, "step": 5850 }, { "epoch": 20.614035087719298, "grad_norm": 0.18243364989757538, "learning_rate": 9.473684210526316e-07, "loss": 0.0015, "step": 5875 }, { "epoch": 20.70175438596491, "grad_norm": 0.14108963310718536, "learning_rate": 8.875598086124402e-07, "loss": 0.0817, "step": 5900 }, { "epoch": 20.789473684210527, "grad_norm": 0.03629032149910927, "learning_rate": 8.277511961722487e-07, "loss": 0.002, "step": 5925 }, { "epoch": 20.87719298245614, "grad_norm": 21.195932388305664, "learning_rate": 7.679425837320574e-07, "loss": 0.1265, "step": 5950 }, { "epoch": 20.964912280701753, "grad_norm": 0.036704275757074356, "learning_rate": 7.081339712918661e-07, "loss": 0.0369, "step": 5975 }, { "epoch": 21.05263157894737, "grad_norm": 327.310791015625, "learning_rate": 6.483253588516747e-07, "loss": 0.0288, "step": 6000 } ], "logging_steps": 25, "max_steps": 6270, "num_input_tokens_seen": 0, "num_train_epochs": 22, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1.6186141004321981e+18, "train_batch_size": 4, "trial_name": null, "trial_params": null }