{ "best_metric": null, "best_model_checkpoint": null, "epoch": 10.0, "eval_steps": 500, "global_step": 114000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.043859649122807015, "grad_norm": 18.61533546447754, "learning_rate": 1.9912280701754387e-05, "loss": 1.5692, "step": 500 }, { "epoch": 0.08771929824561403, "grad_norm": 0.391161173582077, "learning_rate": 1.9824561403508773e-05, "loss": 0.0755, "step": 1000 }, { "epoch": 0.13157894736842105, "grad_norm": 0.4991099238395691, "learning_rate": 1.9736842105263158e-05, "loss": 0.0622, "step": 1500 }, { "epoch": 0.17543859649122806, "grad_norm": 0.22268570959568024, "learning_rate": 1.9649122807017544e-05, "loss": 0.0583, "step": 2000 }, { "epoch": 0.21929824561403508, "grad_norm": 0.6931657195091248, "learning_rate": 1.9561403508771933e-05, "loss": 0.0551, "step": 2500 }, { "epoch": 0.2631578947368421, "grad_norm": 0.6016309261322021, "learning_rate": 1.9473684210526318e-05, "loss": 0.0539, "step": 3000 }, { "epoch": 0.30701754385964913, "grad_norm": 0.46937015652656555, "learning_rate": 1.9385964912280704e-05, "loss": 0.0524, "step": 3500 }, { "epoch": 0.3508771929824561, "grad_norm": 0.2826986610889435, "learning_rate": 1.929824561403509e-05, "loss": 0.0515, "step": 4000 }, { "epoch": 0.39473684210526316, "grad_norm": 0.3649154305458069, "learning_rate": 1.9210526315789474e-05, "loss": 0.0507, "step": 4500 }, { "epoch": 0.43859649122807015, "grad_norm": 2.390549421310425, "learning_rate": 1.912280701754386e-05, "loss": 0.0502, "step": 5000 }, { "epoch": 0.4824561403508772, "grad_norm": 0.8074614405632019, "learning_rate": 1.9035087719298245e-05, "loss": 0.0495, "step": 5500 }, { "epoch": 0.5263157894736842, "grad_norm": 1.1913889646530151, "learning_rate": 1.894736842105263e-05, "loss": 0.0493, "step": 6000 }, { "epoch": 0.5701754385964912, "grad_norm": 0.541630744934082, "learning_rate": 1.885964912280702e-05, "loss": 0.0473, "step": 6500 }, { "epoch": 0.6140350877192983, "grad_norm": 0.4298405051231384, "learning_rate": 1.8771929824561405e-05, "loss": 0.047, "step": 7000 }, { "epoch": 0.6578947368421053, "grad_norm": 0.3407554626464844, "learning_rate": 1.868421052631579e-05, "loss": 0.0465, "step": 7500 }, { "epoch": 0.7017543859649122, "grad_norm": 0.6543895602226257, "learning_rate": 1.8596491228070176e-05, "loss": 0.0449, "step": 8000 }, { "epoch": 0.7456140350877193, "grad_norm": 0.7753346562385559, "learning_rate": 1.8508771929824562e-05, "loss": 0.0446, "step": 8500 }, { "epoch": 0.7894736842105263, "grad_norm": 0.4827066659927368, "learning_rate": 1.8421052631578947e-05, "loss": 0.0448, "step": 9000 }, { "epoch": 0.8333333333333334, "grad_norm": 1.1305031776428223, "learning_rate": 1.8333333333333333e-05, "loss": 0.0436, "step": 9500 }, { "epoch": 0.8771929824561403, "grad_norm": 0.8072851300239563, "learning_rate": 1.824561403508772e-05, "loss": 0.0423, "step": 10000 }, { "epoch": 0.9210526315789473, "grad_norm": 0.693408727645874, "learning_rate": 1.8157894736842107e-05, "loss": 0.0428, "step": 10500 }, { "epoch": 0.9649122807017544, "grad_norm": 0.5725772976875305, "learning_rate": 1.8070175438596493e-05, "loss": 0.0422, "step": 11000 }, { "epoch": 1.0087719298245614, "grad_norm": 0.8863518834114075, "learning_rate": 1.7982456140350878e-05, "loss": 0.04, "step": 11500 }, { "epoch": 1.0526315789473684, "grad_norm": 1.0276522636413574, "learning_rate": 1.7894736842105264e-05, "loss": 0.0366, "step": 12000 }, { "epoch": 1.0964912280701755, "grad_norm": 0.7939022779464722, "learning_rate": 1.780701754385965e-05, "loss": 0.0365, "step": 12500 }, { "epoch": 1.1403508771929824, "grad_norm": 0.9640121459960938, "learning_rate": 1.7719298245614035e-05, "loss": 0.0363, "step": 13000 }, { "epoch": 1.1842105263157894, "grad_norm": 0.9494159817695618, "learning_rate": 1.763157894736842e-05, "loss": 0.0361, "step": 13500 }, { "epoch": 1.2280701754385965, "grad_norm": 1.103772521018982, "learning_rate": 1.754385964912281e-05, "loss": 0.0366, "step": 14000 }, { "epoch": 1.2719298245614035, "grad_norm": 1.3148542642593384, "learning_rate": 1.7456140350877195e-05, "loss": 0.0363, "step": 14500 }, { "epoch": 1.3157894736842106, "grad_norm": 0.9196112751960754, "learning_rate": 1.736842105263158e-05, "loss": 0.0352, "step": 15000 }, { "epoch": 1.3596491228070176, "grad_norm": 0.7027790546417236, "learning_rate": 1.7280701754385966e-05, "loss": 0.0353, "step": 15500 }, { "epoch": 1.4035087719298245, "grad_norm": 0.9489392638206482, "learning_rate": 1.719298245614035e-05, "loss": 0.0348, "step": 16000 }, { "epoch": 1.4473684210526316, "grad_norm": 0.7640049457550049, "learning_rate": 1.7105263157894737e-05, "loss": 0.0347, "step": 16500 }, { "epoch": 1.4912280701754386, "grad_norm": 0.9464479088783264, "learning_rate": 1.7017543859649125e-05, "loss": 0.0347, "step": 17000 }, { "epoch": 1.5350877192982457, "grad_norm": 0.6414206027984619, "learning_rate": 1.692982456140351e-05, "loss": 0.0349, "step": 17500 }, { "epoch": 1.5789473684210527, "grad_norm": 1.0872447490692139, "learning_rate": 1.6842105263157896e-05, "loss": 0.0347, "step": 18000 }, { "epoch": 1.6228070175438596, "grad_norm": 0.8262595534324646, "learning_rate": 1.6754385964912282e-05, "loss": 0.0342, "step": 18500 }, { "epoch": 1.6666666666666665, "grad_norm": 0.8161122798919678, "learning_rate": 1.6666666666666667e-05, "loss": 0.0339, "step": 19000 }, { "epoch": 1.7105263157894737, "grad_norm": 1.061262607574463, "learning_rate": 1.6578947368421053e-05, "loss": 0.0327, "step": 19500 }, { "epoch": 1.7543859649122808, "grad_norm": 0.9424170255661011, "learning_rate": 1.649122807017544e-05, "loss": 0.0332, "step": 20000 }, { "epoch": 1.7982456140350878, "grad_norm": 0.9701814651489258, "learning_rate": 1.6403508771929827e-05, "loss": 0.0339, "step": 20500 }, { "epoch": 1.8421052631578947, "grad_norm": 1.1466444730758667, "learning_rate": 1.6315789473684213e-05, "loss": 0.0323, "step": 21000 }, { "epoch": 1.8859649122807016, "grad_norm": 1.455343246459961, "learning_rate": 1.62280701754386e-05, "loss": 0.0318, "step": 21500 }, { "epoch": 1.9298245614035088, "grad_norm": 0.7721505165100098, "learning_rate": 1.6140350877192984e-05, "loss": 0.0325, "step": 22000 }, { "epoch": 1.973684210526316, "grad_norm": 0.54636150598526, "learning_rate": 1.605263157894737e-05, "loss": 0.0322, "step": 22500 }, { "epoch": 2.017543859649123, "grad_norm": 1.0513033866882324, "learning_rate": 1.5964912280701755e-05, "loss": 0.029, "step": 23000 }, { "epoch": 2.06140350877193, "grad_norm": 1.556970238685608, "learning_rate": 1.5877192982456144e-05, "loss": 0.0261, "step": 23500 }, { "epoch": 2.1052631578947367, "grad_norm": 0.7573184370994568, "learning_rate": 1.578947368421053e-05, "loss": 0.0267, "step": 24000 }, { "epoch": 2.1491228070175437, "grad_norm": 0.929409921169281, "learning_rate": 1.5701754385964915e-05, "loss": 0.0269, "step": 24500 }, { "epoch": 2.192982456140351, "grad_norm": 1.4076780080795288, "learning_rate": 1.56140350877193e-05, "loss": 0.0261, "step": 25000 }, { "epoch": 2.236842105263158, "grad_norm": 1.3167903423309326, "learning_rate": 1.5526315789473686e-05, "loss": 0.0263, "step": 25500 }, { "epoch": 2.280701754385965, "grad_norm": 0.882624626159668, "learning_rate": 1.543859649122807e-05, "loss": 0.0271, "step": 26000 }, { "epoch": 2.324561403508772, "grad_norm": 2.1685619354248047, "learning_rate": 1.535087719298246e-05, "loss": 0.0272, "step": 26500 }, { "epoch": 2.3684210526315788, "grad_norm": 0.8032916784286499, "learning_rate": 1.5263157894736846e-05, "loss": 0.0267, "step": 27000 }, { "epoch": 2.412280701754386, "grad_norm": 0.784699559211731, "learning_rate": 1.517543859649123e-05, "loss": 0.0269, "step": 27500 }, { "epoch": 2.456140350877193, "grad_norm": 1.0281327962875366, "learning_rate": 1.5087719298245615e-05, "loss": 0.027, "step": 28000 }, { "epoch": 2.5, "grad_norm": 0.4517708718776703, "learning_rate": 1.5000000000000002e-05, "loss": 0.0269, "step": 28500 }, { "epoch": 2.543859649122807, "grad_norm": 1.3572579622268677, "learning_rate": 1.4912280701754388e-05, "loss": 0.0267, "step": 29000 }, { "epoch": 2.587719298245614, "grad_norm": 0.673574686050415, "learning_rate": 1.4824561403508773e-05, "loss": 0.0264, "step": 29500 }, { "epoch": 2.6315789473684212, "grad_norm": 1.120792031288147, "learning_rate": 1.4736842105263159e-05, "loss": 0.0277, "step": 30000 }, { "epoch": 2.675438596491228, "grad_norm": 1.4167113304138184, "learning_rate": 1.4649122807017544e-05, "loss": 0.0262, "step": 30500 }, { "epoch": 2.719298245614035, "grad_norm": 0.5733422040939331, "learning_rate": 1.4561403508771931e-05, "loss": 0.0274, "step": 31000 }, { "epoch": 2.763157894736842, "grad_norm": 0.5289915800094604, "learning_rate": 1.4473684210526317e-05, "loss": 0.0265, "step": 31500 }, { "epoch": 2.807017543859649, "grad_norm": 1.224759817123413, "learning_rate": 1.4385964912280704e-05, "loss": 0.0262, "step": 32000 }, { "epoch": 2.8508771929824563, "grad_norm": 0.45227062702178955, "learning_rate": 1.429824561403509e-05, "loss": 0.0264, "step": 32500 }, { "epoch": 2.8947368421052633, "grad_norm": 1.478962779045105, "learning_rate": 1.4210526315789475e-05, "loss": 0.0261, "step": 33000 }, { "epoch": 2.93859649122807, "grad_norm": 0.8974044322967529, "learning_rate": 1.412280701754386e-05, "loss": 0.0268, "step": 33500 }, { "epoch": 2.982456140350877, "grad_norm": 0.9807118773460388, "learning_rate": 1.4035087719298246e-05, "loss": 0.0253, "step": 34000 }, { "epoch": 3.026315789473684, "grad_norm": 1.163085699081421, "learning_rate": 1.3947368421052631e-05, "loss": 0.0215, "step": 34500 }, { "epoch": 3.0701754385964914, "grad_norm": 0.7529701590538025, "learning_rate": 1.385964912280702e-05, "loss": 0.0186, "step": 35000 }, { "epoch": 3.1140350877192984, "grad_norm": 2.0732076168060303, "learning_rate": 1.3771929824561406e-05, "loss": 0.0197, "step": 35500 }, { "epoch": 3.1578947368421053, "grad_norm": 0.7226047515869141, "learning_rate": 1.3684210526315791e-05, "loss": 0.0201, "step": 36000 }, { "epoch": 3.2017543859649122, "grad_norm": 1.686123251914978, "learning_rate": 1.3596491228070177e-05, "loss": 0.0197, "step": 36500 }, { "epoch": 3.245614035087719, "grad_norm": 0.9627934098243713, "learning_rate": 1.3508771929824562e-05, "loss": 0.0194, "step": 37000 }, { "epoch": 3.2894736842105265, "grad_norm": 0.9055445790290833, "learning_rate": 1.3421052631578948e-05, "loss": 0.0205, "step": 37500 }, { "epoch": 3.3333333333333335, "grad_norm": 0.8586304187774658, "learning_rate": 1.3333333333333333e-05, "loss": 0.0202, "step": 38000 }, { "epoch": 3.3771929824561404, "grad_norm": 0.4082282483577728, "learning_rate": 1.3245614035087719e-05, "loss": 0.0202, "step": 38500 }, { "epoch": 3.4210526315789473, "grad_norm": 0.9686273336410522, "learning_rate": 1.3157894736842108e-05, "loss": 0.0208, "step": 39000 }, { "epoch": 3.4649122807017543, "grad_norm": 1.0885154008865356, "learning_rate": 1.3070175438596493e-05, "loss": 0.0201, "step": 39500 }, { "epoch": 3.5087719298245617, "grad_norm": 0.7704782485961914, "learning_rate": 1.2982456140350879e-05, "loss": 0.0209, "step": 40000 }, { "epoch": 3.5526315789473686, "grad_norm": 0.7897283434867859, "learning_rate": 1.2894736842105264e-05, "loss": 0.0207, "step": 40500 }, { "epoch": 3.5964912280701755, "grad_norm": 0.33727455139160156, "learning_rate": 1.280701754385965e-05, "loss": 0.0209, "step": 41000 }, { "epoch": 3.6403508771929824, "grad_norm": 1.086599588394165, "learning_rate": 1.2719298245614035e-05, "loss": 0.021, "step": 41500 }, { "epoch": 3.6842105263157894, "grad_norm": 0.8630421757698059, "learning_rate": 1.263157894736842e-05, "loss": 0.0203, "step": 42000 }, { "epoch": 3.7280701754385968, "grad_norm": 1.400107979774475, "learning_rate": 1.2543859649122808e-05, "loss": 0.0201, "step": 42500 }, { "epoch": 3.7719298245614032, "grad_norm": 0.935147762298584, "learning_rate": 1.2456140350877195e-05, "loss": 0.02, "step": 43000 }, { "epoch": 3.8157894736842106, "grad_norm": 0.6515530347824097, "learning_rate": 1.236842105263158e-05, "loss": 0.0207, "step": 43500 }, { "epoch": 3.8596491228070176, "grad_norm": 1.6099826097488403, "learning_rate": 1.2280701754385966e-05, "loss": 0.0213, "step": 44000 }, { "epoch": 3.9035087719298245, "grad_norm": 1.4369622468948364, "learning_rate": 1.2192982456140352e-05, "loss": 0.0198, "step": 44500 }, { "epoch": 3.9473684210526314, "grad_norm": 0.46659207344055176, "learning_rate": 1.2105263157894737e-05, "loss": 0.0211, "step": 45000 }, { "epoch": 3.9912280701754383, "grad_norm": 1.361497402191162, "learning_rate": 1.2017543859649124e-05, "loss": 0.0208, "step": 45500 }, { "epoch": 4.035087719298246, "grad_norm": 0.664008378982544, "learning_rate": 1.192982456140351e-05, "loss": 0.0148, "step": 46000 }, { "epoch": 4.078947368421052, "grad_norm": 0.9905796647071838, "learning_rate": 1.1842105263157895e-05, "loss": 0.0132, "step": 46500 }, { "epoch": 4.12280701754386, "grad_norm": 0.9070369601249695, "learning_rate": 1.1754385964912282e-05, "loss": 0.0134, "step": 47000 }, { "epoch": 4.166666666666667, "grad_norm": 1.3704754114151, "learning_rate": 1.1666666666666668e-05, "loss": 0.0129, "step": 47500 }, { "epoch": 4.2105263157894735, "grad_norm": 2.792363166809082, "learning_rate": 1.1578947368421053e-05, "loss": 0.0141, "step": 48000 }, { "epoch": 4.254385964912281, "grad_norm": 0.19838610291481018, "learning_rate": 1.1491228070175439e-05, "loss": 0.014, "step": 48500 }, { "epoch": 4.298245614035087, "grad_norm": 1.3491791486740112, "learning_rate": 1.1403508771929826e-05, "loss": 0.0145, "step": 49000 }, { "epoch": 4.342105263157895, "grad_norm": 0.7217977046966553, "learning_rate": 1.1315789473684212e-05, "loss": 0.0143, "step": 49500 }, { "epoch": 4.385964912280702, "grad_norm": 1.6904853582382202, "learning_rate": 1.1228070175438597e-05, "loss": 0.015, "step": 50000 }, { "epoch": 4.4298245614035086, "grad_norm": 1.9361881017684937, "learning_rate": 1.1140350877192983e-05, "loss": 0.0151, "step": 50500 }, { "epoch": 4.473684210526316, "grad_norm": 1.5735112428665161, "learning_rate": 1.105263157894737e-05, "loss": 0.015, "step": 51000 }, { "epoch": 4.517543859649123, "grad_norm": 0.5673351287841797, "learning_rate": 1.0964912280701755e-05, "loss": 0.0148, "step": 51500 }, { "epoch": 4.56140350877193, "grad_norm": 0.6633049249649048, "learning_rate": 1.0877192982456142e-05, "loss": 0.0151, "step": 52000 }, { "epoch": 4.605263157894737, "grad_norm": 0.7345007061958313, "learning_rate": 1.0789473684210528e-05, "loss": 0.0151, "step": 52500 }, { "epoch": 4.649122807017544, "grad_norm": 0.44649428129196167, "learning_rate": 1.0701754385964913e-05, "loss": 0.0157, "step": 53000 }, { "epoch": 4.692982456140351, "grad_norm": 0.7061645984649658, "learning_rate": 1.0614035087719299e-05, "loss": 0.0143, "step": 53500 }, { "epoch": 4.7368421052631575, "grad_norm": 0.4091035723686218, "learning_rate": 1.0526315789473684e-05, "loss": 0.015, "step": 54000 }, { "epoch": 4.780701754385965, "grad_norm": 1.0567431449890137, "learning_rate": 1.043859649122807e-05, "loss": 0.0153, "step": 54500 }, { "epoch": 4.824561403508772, "grad_norm": 1.535997748374939, "learning_rate": 1.0350877192982459e-05, "loss": 0.0145, "step": 55000 }, { "epoch": 4.868421052631579, "grad_norm": 2.3148131370544434, "learning_rate": 1.0263157894736844e-05, "loss": 0.0152, "step": 55500 }, { "epoch": 4.912280701754386, "grad_norm": 1.4357719421386719, "learning_rate": 1.017543859649123e-05, "loss": 0.0154, "step": 56000 }, { "epoch": 4.956140350877193, "grad_norm": 2.350245714187622, "learning_rate": 1.0087719298245615e-05, "loss": 0.0153, "step": 56500 }, { "epoch": 5.0, "grad_norm": 1.0698745250701904, "learning_rate": 1e-05, "loss": 0.0154, "step": 57000 }, { "epoch": 5.043859649122807, "grad_norm": 2.993831157684326, "learning_rate": 9.912280701754386e-06, "loss": 0.0077, "step": 57500 }, { "epoch": 5.087719298245614, "grad_norm": 0.41161152720451355, "learning_rate": 9.824561403508772e-06, "loss": 0.0084, "step": 58000 }, { "epoch": 5.131578947368421, "grad_norm": 1.7437654733657837, "learning_rate": 9.736842105263159e-06, "loss": 0.0086, "step": 58500 }, { "epoch": 5.175438596491228, "grad_norm": 0.5381312966346741, "learning_rate": 9.649122807017545e-06, "loss": 0.0083, "step": 59000 }, { "epoch": 5.219298245614035, "grad_norm": 0.15729767084121704, "learning_rate": 9.56140350877193e-06, "loss": 0.0099, "step": 59500 }, { "epoch": 5.2631578947368425, "grad_norm": 1.227547526359558, "learning_rate": 9.473684210526315e-06, "loss": 0.0094, "step": 60000 }, { "epoch": 5.307017543859649, "grad_norm": 1.4580745697021484, "learning_rate": 9.385964912280703e-06, "loss": 0.0101, "step": 60500 }, { "epoch": 5.350877192982456, "grad_norm": 1.4596012830734253, "learning_rate": 9.298245614035088e-06, "loss": 0.009, "step": 61000 }, { "epoch": 5.394736842105263, "grad_norm": 0.3997878432273865, "learning_rate": 9.210526315789474e-06, "loss": 0.0095, "step": 61500 }, { "epoch": 5.43859649122807, "grad_norm": 1.4484952688217163, "learning_rate": 9.12280701754386e-06, "loss": 0.0095, "step": 62000 }, { "epoch": 5.482456140350878, "grad_norm": 0.3741638958454132, "learning_rate": 9.035087719298246e-06, "loss": 0.0095, "step": 62500 }, { "epoch": 5.526315789473684, "grad_norm": 1.792814016342163, "learning_rate": 8.947368421052632e-06, "loss": 0.0103, "step": 63000 }, { "epoch": 5.5701754385964914, "grad_norm": 1.4373083114624023, "learning_rate": 8.859649122807017e-06, "loss": 0.01, "step": 63500 }, { "epoch": 5.614035087719298, "grad_norm": 0.513109564781189, "learning_rate": 8.771929824561405e-06, "loss": 0.0095, "step": 64000 }, { "epoch": 5.657894736842105, "grad_norm": 1.6872016191482544, "learning_rate": 8.68421052631579e-06, "loss": 0.0096, "step": 64500 }, { "epoch": 5.701754385964913, "grad_norm": 1.5767537355422974, "learning_rate": 8.596491228070176e-06, "loss": 0.0095, "step": 65000 }, { "epoch": 5.745614035087719, "grad_norm": 3.215756893157959, "learning_rate": 8.508771929824563e-06, "loss": 0.0104, "step": 65500 }, { "epoch": 5.7894736842105265, "grad_norm": 0.3637864589691162, "learning_rate": 8.421052631578948e-06, "loss": 0.0097, "step": 66000 }, { "epoch": 5.833333333333333, "grad_norm": 3.017098903656006, "learning_rate": 8.333333333333334e-06, "loss": 0.0104, "step": 66500 }, { "epoch": 5.87719298245614, "grad_norm": 0.7917022705078125, "learning_rate": 8.24561403508772e-06, "loss": 0.0105, "step": 67000 }, { "epoch": 5.921052631578947, "grad_norm": 0.28688350319862366, "learning_rate": 8.157894736842106e-06, "loss": 0.0097, "step": 67500 }, { "epoch": 5.964912280701754, "grad_norm": 0.6055354475975037, "learning_rate": 8.070175438596492e-06, "loss": 0.0102, "step": 68000 }, { "epoch": 6.008771929824562, "grad_norm": 0.01086004264652729, "learning_rate": 7.982456140350877e-06, "loss": 0.0092, "step": 68500 }, { "epoch": 6.052631578947368, "grad_norm": 0.53713458776474, "learning_rate": 7.894736842105265e-06, "loss": 0.0049, "step": 69000 }, { "epoch": 6.0964912280701755, "grad_norm": 0.6804053783416748, "learning_rate": 7.80701754385965e-06, "loss": 0.0054, "step": 69500 }, { "epoch": 6.140350877192983, "grad_norm": 0.3400014042854309, "learning_rate": 7.719298245614036e-06, "loss": 0.0053, "step": 70000 }, { "epoch": 6.184210526315789, "grad_norm": 1.194455862045288, "learning_rate": 7.631578947368423e-06, "loss": 0.0052, "step": 70500 }, { "epoch": 6.228070175438597, "grad_norm": 0.1484297811985016, "learning_rate": 7.5438596491228074e-06, "loss": 0.0062, "step": 71000 }, { "epoch": 6.271929824561403, "grad_norm": 1.752806305885315, "learning_rate": 7.456140350877194e-06, "loss": 0.0056, "step": 71500 }, { "epoch": 6.315789473684211, "grad_norm": 0.23840796947479248, "learning_rate": 7.368421052631579e-06, "loss": 0.0057, "step": 72000 }, { "epoch": 6.359649122807017, "grad_norm": 0.6682614684104919, "learning_rate": 7.280701754385966e-06, "loss": 0.0058, "step": 72500 }, { "epoch": 6.4035087719298245, "grad_norm": 0.3302547335624695, "learning_rate": 7.192982456140352e-06, "loss": 0.0059, "step": 73000 }, { "epoch": 6.447368421052632, "grad_norm": 2.1811578273773193, "learning_rate": 7.1052631578947375e-06, "loss": 0.0062, "step": 73500 }, { "epoch": 6.491228070175438, "grad_norm": 0.13947592675685883, "learning_rate": 7.017543859649123e-06, "loss": 0.0064, "step": 74000 }, { "epoch": 6.535087719298246, "grad_norm": 0.1955937296152115, "learning_rate": 6.92982456140351e-06, "loss": 0.0059, "step": 74500 }, { "epoch": 6.578947368421053, "grad_norm": 1.8887696266174316, "learning_rate": 6.842105263157896e-06, "loss": 0.006, "step": 75000 }, { "epoch": 6.62280701754386, "grad_norm": 1.8402165174484253, "learning_rate": 6.754385964912281e-06, "loss": 0.0059, "step": 75500 }, { "epoch": 6.666666666666667, "grad_norm": 0.09785465896129608, "learning_rate": 6.666666666666667e-06, "loss": 0.006, "step": 76000 }, { "epoch": 6.7105263157894735, "grad_norm": 0.060049816966056824, "learning_rate": 6.578947368421054e-06, "loss": 0.0067, "step": 76500 }, { "epoch": 6.754385964912281, "grad_norm": 0.9939663410186768, "learning_rate": 6.491228070175439e-06, "loss": 0.0064, "step": 77000 }, { "epoch": 6.798245614035087, "grad_norm": 2.25742769241333, "learning_rate": 6.403508771929825e-06, "loss": 0.0058, "step": 77500 }, { "epoch": 6.842105263157895, "grad_norm": 1.2935279607772827, "learning_rate": 6.31578947368421e-06, "loss": 0.006, "step": 78000 }, { "epoch": 6.885964912280702, "grad_norm": 0.9248834252357483, "learning_rate": 6.2280701754385975e-06, "loss": 0.0063, "step": 78500 }, { "epoch": 6.9298245614035086, "grad_norm": 0.40686044096946716, "learning_rate": 6.140350877192983e-06, "loss": 0.006, "step": 79000 }, { "epoch": 6.973684210526316, "grad_norm": 0.13613936305046082, "learning_rate": 6.0526315789473685e-06, "loss": 0.0065, "step": 79500 }, { "epoch": 7.017543859649122, "grad_norm": 5.229935169219971, "learning_rate": 5.964912280701755e-06, "loss": 0.0051, "step": 80000 }, { "epoch": 7.06140350877193, "grad_norm": 0.18177901208400726, "learning_rate": 5.877192982456141e-06, "loss": 0.0028, "step": 80500 }, { "epoch": 7.105263157894737, "grad_norm": 7.578922748565674, "learning_rate": 5.789473684210527e-06, "loss": 0.0029, "step": 81000 }, { "epoch": 7.149122807017544, "grad_norm": 7.015735149383545, "learning_rate": 5.701754385964913e-06, "loss": 0.003, "step": 81500 }, { "epoch": 7.192982456140351, "grad_norm": 0.00804635975509882, "learning_rate": 5.6140350877192985e-06, "loss": 0.0033, "step": 82000 }, { "epoch": 7.2368421052631575, "grad_norm": 0.20541447401046753, "learning_rate": 5.526315789473685e-06, "loss": 0.0031, "step": 82500 }, { "epoch": 7.280701754385965, "grad_norm": 2.909257173538208, "learning_rate": 5.438596491228071e-06, "loss": 0.0033, "step": 83000 }, { "epoch": 7.324561403508772, "grad_norm": 0.04006243869662285, "learning_rate": 5.350877192982457e-06, "loss": 0.0031, "step": 83500 }, { "epoch": 7.368421052631579, "grad_norm": 0.011600236408412457, "learning_rate": 5.263157894736842e-06, "loss": 0.0032, "step": 84000 }, { "epoch": 7.412280701754386, "grad_norm": 0.10742747783660889, "learning_rate": 5.175438596491229e-06, "loss": 0.0034, "step": 84500 }, { "epoch": 7.456140350877193, "grad_norm": 0.004623486194759607, "learning_rate": 5.087719298245615e-06, "loss": 0.0039, "step": 85000 }, { "epoch": 7.5, "grad_norm": 0.014665776863694191, "learning_rate": 5e-06, "loss": 0.0029, "step": 85500 }, { "epoch": 7.543859649122807, "grad_norm": 3.925727605819702, "learning_rate": 4.912280701754386e-06, "loss": 0.0031, "step": 86000 }, { "epoch": 7.587719298245614, "grad_norm": 4.379343509674072, "learning_rate": 4.824561403508772e-06, "loss": 0.0032, "step": 86500 }, { "epoch": 7.631578947368421, "grad_norm": 0.3804740011692047, "learning_rate": 4.736842105263158e-06, "loss": 0.0032, "step": 87000 }, { "epoch": 7.675438596491228, "grad_norm": 5.974408149719238, "learning_rate": 4.649122807017544e-06, "loss": 0.0037, "step": 87500 }, { "epoch": 7.719298245614035, "grad_norm": 0.14405211806297302, "learning_rate": 4.56140350877193e-06, "loss": 0.0029, "step": 88000 }, { "epoch": 7.7631578947368425, "grad_norm": 1.343384861946106, "learning_rate": 4.473684210526316e-06, "loss": 0.0031, "step": 88500 }, { "epoch": 7.807017543859649, "grad_norm": 1.0002413988113403, "learning_rate": 4.385964912280702e-06, "loss": 0.0035, "step": 89000 }, { "epoch": 7.850877192982456, "grad_norm": 0.08076108992099762, "learning_rate": 4.298245614035088e-06, "loss": 0.0034, "step": 89500 }, { "epoch": 7.894736842105263, "grad_norm": 1.5308058261871338, "learning_rate": 4.210526315789474e-06, "loss": 0.0033, "step": 90000 }, { "epoch": 7.93859649122807, "grad_norm": 0.05943544954061508, "learning_rate": 4.12280701754386e-06, "loss": 0.0034, "step": 90500 }, { "epoch": 7.982456140350877, "grad_norm": 0.04788463935256004, "learning_rate": 4.035087719298246e-06, "loss": 0.0033, "step": 91000 }, { "epoch": 8.026315789473685, "grad_norm": 0.03480690345168114, "learning_rate": 3.947368421052632e-06, "loss": 0.0019, "step": 91500 }, { "epoch": 8.070175438596491, "grad_norm": 0.028245439752936363, "learning_rate": 3.859649122807018e-06, "loss": 0.0013, "step": 92000 }, { "epoch": 8.114035087719298, "grad_norm": 1.6024084091186523, "learning_rate": 3.7719298245614037e-06, "loss": 0.0012, "step": 92500 }, { "epoch": 8.157894736842104, "grad_norm": 0.023393429815769196, "learning_rate": 3.6842105263157896e-06, "loss": 0.001, "step": 93000 }, { "epoch": 8.201754385964913, "grad_norm": 0.0070739006623625755, "learning_rate": 3.596491228070176e-06, "loss": 0.0014, "step": 93500 }, { "epoch": 8.24561403508772, "grad_norm": 0.005052722524851561, "learning_rate": 3.5087719298245615e-06, "loss": 0.0014, "step": 94000 }, { "epoch": 8.289473684210526, "grad_norm": 0.007294114679098129, "learning_rate": 3.421052631578948e-06, "loss": 0.0015, "step": 94500 }, { "epoch": 8.333333333333334, "grad_norm": 0.0006771266926079988, "learning_rate": 3.3333333333333333e-06, "loss": 0.0012, "step": 95000 }, { "epoch": 8.37719298245614, "grad_norm": 0.02214032970368862, "learning_rate": 3.2456140350877197e-06, "loss": 0.0014, "step": 95500 }, { "epoch": 8.421052631578947, "grad_norm": 2.873591899871826, "learning_rate": 3.157894736842105e-06, "loss": 0.0016, "step": 96000 }, { "epoch": 8.464912280701755, "grad_norm": 0.0030384217388927937, "learning_rate": 3.0701754385964915e-06, "loss": 0.0014, "step": 96500 }, { "epoch": 8.508771929824562, "grad_norm": 1.468751311302185, "learning_rate": 2.9824561403508774e-06, "loss": 0.0018, "step": 97000 }, { "epoch": 8.552631578947368, "grad_norm": 0.005693173501640558, "learning_rate": 2.8947368421052634e-06, "loss": 0.0017, "step": 97500 }, { "epoch": 8.596491228070175, "grad_norm": 0.0886327475309372, "learning_rate": 2.8070175438596493e-06, "loss": 0.0013, "step": 98000 }, { "epoch": 8.640350877192983, "grad_norm": 4.932157516479492, "learning_rate": 2.7192982456140356e-06, "loss": 0.0015, "step": 98500 }, { "epoch": 8.68421052631579, "grad_norm": 0.030486512929201126, "learning_rate": 2.631578947368421e-06, "loss": 0.0015, "step": 99000 }, { "epoch": 8.728070175438596, "grad_norm": 0.02273370511829853, "learning_rate": 2.5438596491228075e-06, "loss": 0.0015, "step": 99500 }, { "epoch": 8.771929824561404, "grad_norm": 0.0035624816082417965, "learning_rate": 2.456140350877193e-06, "loss": 0.0015, "step": 100000 }, { "epoch": 8.81578947368421, "grad_norm": 0.07872194796800613, "learning_rate": 2.368421052631579e-06, "loss": 0.0017, "step": 100500 }, { "epoch": 8.859649122807017, "grad_norm": 0.13730837404727936, "learning_rate": 2.280701754385965e-06, "loss": 0.0014, "step": 101000 }, { "epoch": 8.903508771929825, "grad_norm": 0.006974358111619949, "learning_rate": 2.192982456140351e-06, "loss": 0.0013, "step": 101500 }, { "epoch": 8.947368421052632, "grad_norm": 0.005287351552397013, "learning_rate": 2.105263157894737e-06, "loss": 0.0011, "step": 102000 }, { "epoch": 8.991228070175438, "grad_norm": 0.037747837603092194, "learning_rate": 2.017543859649123e-06, "loss": 0.0015, "step": 102500 }, { "epoch": 9.035087719298245, "grad_norm": 0.7193840146064758, "learning_rate": 1.929824561403509e-06, "loss": 0.0005, "step": 103000 }, { "epoch": 9.078947368421053, "grad_norm": 0.0007897180039435625, "learning_rate": 1.8421052631578948e-06, "loss": 0.0004, "step": 103500 }, { "epoch": 9.12280701754386, "grad_norm": 0.0007234093500301242, "learning_rate": 1.7543859649122807e-06, "loss": 0.0005, "step": 104000 }, { "epoch": 9.166666666666666, "grad_norm": 0.0007756888517178595, "learning_rate": 1.6666666666666667e-06, "loss": 0.0003, "step": 104500 }, { "epoch": 9.210526315789474, "grad_norm": 0.008586079813539982, "learning_rate": 1.5789473684210526e-06, "loss": 0.0006, "step": 105000 }, { "epoch": 9.25438596491228, "grad_norm": 0.00045911312918178737, "learning_rate": 1.4912280701754387e-06, "loss": 0.0004, "step": 105500 }, { "epoch": 9.298245614035087, "grad_norm": 0.00909907091408968, "learning_rate": 1.4035087719298246e-06, "loss": 0.0004, "step": 106000 }, { "epoch": 9.342105263157896, "grad_norm": 0.0010117848869413137, "learning_rate": 1.3157894736842106e-06, "loss": 0.0005, "step": 106500 }, { "epoch": 9.385964912280702, "grad_norm": 0.06939810514450073, "learning_rate": 1.2280701754385965e-06, "loss": 0.0004, "step": 107000 }, { "epoch": 9.429824561403509, "grad_norm": 0.00043201466905884445, "learning_rate": 1.1403508771929824e-06, "loss": 0.0005, "step": 107500 }, { "epoch": 9.473684210526315, "grad_norm": 0.0020823481027036905, "learning_rate": 1.0526315789473685e-06, "loss": 0.0003, "step": 108000 }, { "epoch": 9.517543859649123, "grad_norm": 0.001455445308238268, "learning_rate": 9.649122807017545e-07, "loss": 0.0004, "step": 108500 }, { "epoch": 9.56140350877193, "grad_norm": 0.04049188643693924, "learning_rate": 8.771929824561404e-07, "loss": 0.0004, "step": 109000 }, { "epoch": 9.605263157894736, "grad_norm": 2.738844156265259, "learning_rate": 7.894736842105263e-07, "loss": 0.0005, "step": 109500 }, { "epoch": 9.649122807017545, "grad_norm": 0.0004239229310769588, "learning_rate": 7.017543859649123e-07, "loss": 0.0004, "step": 110000 }, { "epoch": 9.692982456140351, "grad_norm": 0.012256275862455368, "learning_rate": 6.140350877192982e-07, "loss": 0.0004, "step": 110500 }, { "epoch": 9.736842105263158, "grad_norm": 0.004417608492076397, "learning_rate": 5.263157894736843e-07, "loss": 0.0004, "step": 111000 }, { "epoch": 9.780701754385966, "grad_norm": 0.04168286547064781, "learning_rate": 4.385964912280702e-07, "loss": 0.0003, "step": 111500 }, { "epoch": 9.824561403508772, "grad_norm": 0.00598061503842473, "learning_rate": 3.5087719298245616e-07, "loss": 0.0003, "step": 112000 }, { "epoch": 9.868421052631579, "grad_norm": 0.0009545275825075805, "learning_rate": 2.6315789473684213e-07, "loss": 0.0007, "step": 112500 }, { "epoch": 9.912280701754385, "grad_norm": 0.004537086468189955, "learning_rate": 1.7543859649122808e-07, "loss": 0.0004, "step": 113000 }, { "epoch": 9.956140350877194, "grad_norm": 0.0037381595466285944, "learning_rate": 8.771929824561404e-08, "loss": 0.0005, "step": 113500 }, { "epoch": 10.0, "grad_norm": 0.0002570266369730234, "learning_rate": 0.0, "loss": 0.0003, "step": 114000 } ], "logging_steps": 500, "max_steps": 114000, "num_input_tokens_seen": 0, "num_train_epochs": 10, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 2.128164391181353e+18, "train_batch_size": 24, "trial_name": null, "trial_params": null }