{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.0010590639992374739, "eval_steps": 9, "global_step": 50, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 2.118127998474948e-05, "grad_norm": 0.16889730095863342, "learning_rate": 1e-05, "loss": 10.3956, "step": 1 }, { "epoch": 2.118127998474948e-05, "eval_loss": 10.398905754089355, "eval_runtime": 263.6586, "eval_samples_per_second": 150.794, "eval_steps_per_second": 18.85, "step": 1 }, { "epoch": 4.236255996949896e-05, "grad_norm": 0.17050734162330627, "learning_rate": 2e-05, "loss": 10.3876, "step": 2 }, { "epoch": 6.354383995424843e-05, "grad_norm": 0.1798275262117386, "learning_rate": 3e-05, "loss": 10.3979, "step": 3 }, { "epoch": 8.472511993899791e-05, "grad_norm": 0.1841912865638733, "learning_rate": 4e-05, "loss": 10.3896, "step": 4 }, { "epoch": 0.0001059063999237474, "grad_norm": 0.18602658808231354, "learning_rate": 5e-05, "loss": 10.4073, "step": 5 }, { "epoch": 0.00012708767990849687, "grad_norm": 0.1590508371591568, "learning_rate": 6e-05, "loss": 10.408, "step": 6 }, { "epoch": 0.00014826895989324636, "grad_norm": 0.18670088052749634, "learning_rate": 7e-05, "loss": 10.3834, "step": 7 }, { "epoch": 0.00016945023987799583, "grad_norm": 0.24394260346889496, "learning_rate": 8e-05, "loss": 10.3698, "step": 8 }, { "epoch": 0.0001906315198627453, "grad_norm": 0.18870773911476135, "learning_rate": 9e-05, "loss": 10.4098, "step": 9 }, { "epoch": 0.0001906315198627453, "eval_loss": 10.396601676940918, "eval_runtime": 264.5263, "eval_samples_per_second": 150.299, "eval_steps_per_second": 18.788, "step": 9 }, { "epoch": 0.0002118127998474948, "grad_norm": 0.18838222324848175, "learning_rate": 0.0001, "loss": 10.3923, "step": 10 }, { "epoch": 0.00023299407983224426, "grad_norm": 0.16910575330257416, "learning_rate": 9.99695413509548e-05, "loss": 10.3948, "step": 11 }, { "epoch": 0.00025417535981699373, "grad_norm": 0.19666463136672974, "learning_rate": 9.987820251299122e-05, "loss": 10.3688, "step": 12 }, { "epoch": 0.0002753566398017432, "grad_norm": 0.17982766032218933, "learning_rate": 9.972609476841367e-05, "loss": 10.3875, "step": 13 }, { "epoch": 0.0002965379197864927, "grad_norm": 0.19756174087524414, "learning_rate": 9.951340343707852e-05, "loss": 10.3725, "step": 14 }, { "epoch": 0.0003177191997712422, "grad_norm": 0.21875005960464478, "learning_rate": 9.924038765061042e-05, "loss": 10.3732, "step": 15 }, { "epoch": 0.00033890047975599166, "grad_norm": 0.22114615142345428, "learning_rate": 9.890738003669029e-05, "loss": 10.401, "step": 16 }, { "epoch": 0.00036008175974074113, "grad_norm": 0.18753595650196075, "learning_rate": 9.851478631379982e-05, "loss": 10.3818, "step": 17 }, { "epoch": 0.0003812630397254906, "grad_norm": 0.1712396889925003, "learning_rate": 9.806308479691595e-05, "loss": 10.3843, "step": 18 }, { "epoch": 0.0003812630397254906, "eval_loss": 10.390729904174805, "eval_runtime": 264.2162, "eval_samples_per_second": 150.475, "eval_steps_per_second": 18.81, "step": 18 }, { "epoch": 0.00040244431971024006, "grad_norm": 0.22963330149650574, "learning_rate": 9.755282581475769e-05, "loss": 10.3631, "step": 19 }, { "epoch": 0.0004236255996949896, "grad_norm": 0.18184514343738556, "learning_rate": 9.698463103929542e-05, "loss": 10.3788, "step": 20 }, { "epoch": 0.00044480687967973906, "grad_norm": 0.19332440197467804, "learning_rate": 9.635919272833938e-05, "loss": 10.3804, "step": 21 }, { "epoch": 0.0004659881596644885, "grad_norm": 0.19627565145492554, "learning_rate": 9.567727288213005e-05, "loss": 10.3853, "step": 22 }, { "epoch": 0.000487169439649238, "grad_norm": 0.19879043102264404, "learning_rate": 9.493970231495835e-05, "loss": 10.3913, "step": 23 }, { "epoch": 0.0005083507196339875, "grad_norm": 0.19176159799098969, "learning_rate": 9.414737964294636e-05, "loss": 10.3894, "step": 24 }, { "epoch": 0.0005295319996187369, "grad_norm": 0.22696080803871155, "learning_rate": 9.330127018922194e-05, "loss": 10.3604, "step": 25 }, { "epoch": 0.0005507132796034864, "grad_norm": 0.20836742222309113, "learning_rate": 9.24024048078213e-05, "loss": 10.3644, "step": 26 }, { "epoch": 0.0005718945595882359, "grad_norm": 0.2002551406621933, "learning_rate": 9.145187862775209e-05, "loss": 10.4058, "step": 27 }, { "epoch": 0.0005718945595882359, "eval_loss": 10.384385108947754, "eval_runtime": 264.2193, "eval_samples_per_second": 150.473, "eval_steps_per_second": 18.81, "step": 27 }, { "epoch": 0.0005930758395729854, "grad_norm": 0.2085038721561432, "learning_rate": 9.045084971874738e-05, "loss": 10.375, "step": 28 }, { "epoch": 0.0006142571195577349, "grad_norm": 0.25786292552948, "learning_rate": 8.940053768033609e-05, "loss": 10.4152, "step": 29 }, { "epoch": 0.0006354383995424844, "grad_norm": 0.2436630129814148, "learning_rate": 8.83022221559489e-05, "loss": 10.3652, "step": 30 }, { "epoch": 0.0006566196795272338, "grad_norm": 0.20807240903377533, "learning_rate": 8.715724127386972e-05, "loss": 10.3717, "step": 31 }, { "epoch": 0.0006778009595119833, "grad_norm": 0.21431414783000946, "learning_rate": 8.596699001693255e-05, "loss": 10.3851, "step": 32 }, { "epoch": 0.0006989822394967328, "grad_norm": 0.2529206871986389, "learning_rate": 8.473291852294987e-05, "loss": 10.3919, "step": 33 }, { "epoch": 0.0007201635194814823, "grad_norm": 0.20583346486091614, "learning_rate": 8.345653031794292e-05, "loss": 10.3932, "step": 34 }, { "epoch": 0.0007413447994662317, "grad_norm": 0.19697058200836182, "learning_rate": 8.213938048432697e-05, "loss": 10.3836, "step": 35 }, { "epoch": 0.0007625260794509812, "grad_norm": 0.22605247795581818, "learning_rate": 8.07830737662829e-05, "loss": 10.3709, "step": 36 }, { "epoch": 0.0007625260794509812, "eval_loss": 10.377544403076172, "eval_runtime": 264.2986, "eval_samples_per_second": 150.428, "eval_steps_per_second": 18.804, "step": 36 }, { "epoch": 0.0007837073594357307, "grad_norm": 0.23540405929088593, "learning_rate": 7.938926261462366e-05, "loss": 10.3821, "step": 37 }, { "epoch": 0.0008048886394204801, "grad_norm": 0.25848376750946045, "learning_rate": 7.795964517353735e-05, "loss": 10.3993, "step": 38 }, { "epoch": 0.0008260699194052297, "grad_norm": 0.2196625918149948, "learning_rate": 7.649596321166024e-05, "loss": 10.3743, "step": 39 }, { "epoch": 0.0008472511993899792, "grad_norm": 0.22024263441562653, "learning_rate": 7.500000000000001e-05, "loss": 10.3815, "step": 40 }, { "epoch": 0.0008684324793747286, "grad_norm": 0.22341415286064148, "learning_rate": 7.347357813929454e-05, "loss": 10.4108, "step": 41 }, { "epoch": 0.0008896137593594781, "grad_norm": 0.2678864300251007, "learning_rate": 7.191855733945387e-05, "loss": 10.3734, "step": 42 }, { "epoch": 0.0009107950393442276, "grad_norm": 0.24385559558868408, "learning_rate": 7.033683215379002e-05, "loss": 10.3607, "step": 43 }, { "epoch": 0.000931976319328977, "grad_norm": 0.2389352172613144, "learning_rate": 6.873032967079561e-05, "loss": 10.36, "step": 44 }, { "epoch": 0.0009531575993137265, "grad_norm": 0.24411995708942413, "learning_rate": 6.710100716628344e-05, "loss": 10.3555, "step": 45 }, { "epoch": 0.0009531575993137265, "eval_loss": 10.370607376098633, "eval_runtime": 264.3906, "eval_samples_per_second": 150.376, "eval_steps_per_second": 18.798, "step": 45 }, { "epoch": 0.000974338879298476, "grad_norm": 0.24778561294078827, "learning_rate": 6.545084971874738e-05, "loss": 10.3654, "step": 46 }, { "epoch": 0.0009955201592832255, "grad_norm": 0.27637922763824463, "learning_rate": 6.378186779084995e-05, "loss": 10.3488, "step": 47 }, { "epoch": 0.001016701439267975, "grad_norm": 0.28022265434265137, "learning_rate": 6.209609477998338e-05, "loss": 10.3677, "step": 48 }, { "epoch": 0.0010378827192527244, "grad_norm": 0.2709881365299225, "learning_rate": 6.0395584540887963e-05, "loss": 10.349, "step": 49 }, { "epoch": 0.0010590639992374739, "grad_norm": 0.2436312437057495, "learning_rate": 5.868240888334653e-05, "loss": 10.3602, "step": 50 } ], "logging_steps": 1, "max_steps": 100, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 25, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 2615122329600.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }