| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.47368421052631576, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004784688995215311, | |
| "eval_loss": 10.365926742553711, | |
| "eval_runtime": 1.8443, | |
| "eval_samples_per_second": 762.886, | |
| "eval_steps_per_second": 23.857, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014354066985645933, | |
| "grad_norm": 0.1034882590174675, | |
| "learning_rate": 3e-05, | |
| "loss": 10.3664, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.028708133971291867, | |
| "grad_norm": 0.10635757446289062, | |
| "learning_rate": 6e-05, | |
| "loss": 10.3653, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0430622009569378, | |
| "grad_norm": 0.11460530757904053, | |
| "learning_rate": 9e-05, | |
| "loss": 10.3645, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0430622009569378, | |
| "eval_loss": 10.362683296203613, | |
| "eval_runtime": 1.8443, | |
| "eval_samples_per_second": 762.883, | |
| "eval_steps_per_second": 23.857, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05741626794258373, | |
| "grad_norm": 0.12390146404504776, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 10.3624, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.07177033492822966, | |
| "grad_norm": 0.14381946623325348, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 10.3584, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0861244019138756, | |
| "grad_norm": 0.16828647255897522, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 10.3555, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0861244019138756, | |
| "eval_loss": 10.352652549743652, | |
| "eval_runtime": 1.87, | |
| "eval_samples_per_second": 752.399, | |
| "eval_steps_per_second": 23.529, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10047846889952153, | |
| "grad_norm": 0.19975420832633972, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 10.3511, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.11483253588516747, | |
| "grad_norm": 0.21127574145793915, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 10.3466, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1291866028708134, | |
| "grad_norm": 0.21262452006340027, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 10.3415, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1291866028708134, | |
| "eval_loss": 10.338648796081543, | |
| "eval_runtime": 1.8449, | |
| "eval_samples_per_second": 762.643, | |
| "eval_steps_per_second": 23.85, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.14354066985645933, | |
| "grad_norm": 0.20146459341049194, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 10.3382, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15789473684210525, | |
| "grad_norm": 0.1689472794532776, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 10.3341, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1722488038277512, | |
| "grad_norm": 0.1589594930410385, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 10.3297, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1722488038277512, | |
| "eval_loss": 10.327372550964355, | |
| "eval_runtime": 1.8458, | |
| "eval_samples_per_second": 762.288, | |
| "eval_steps_per_second": 23.838, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.18660287081339713, | |
| "grad_norm": 0.14408323168754578, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 10.3273, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.20095693779904306, | |
| "grad_norm": 0.13509996235370636, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 10.3244, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.215311004784689, | |
| "grad_norm": 0.11339321732521057, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 10.3244, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.215311004784689, | |
| "eval_loss": 10.321708679199219, | |
| "eval_runtime": 1.8345, | |
| "eval_samples_per_second": 766.961, | |
| "eval_steps_per_second": 23.985, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.22966507177033493, | |
| "grad_norm": 0.108404740691185, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 10.3215, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.24401913875598086, | |
| "grad_norm": 0.10472545027732849, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 10.3201, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.2583732057416268, | |
| "grad_norm": 0.10555419325828552, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 10.3182, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2583732057416268, | |
| "eval_loss": 10.317863464355469, | |
| "eval_runtime": 1.8688, | |
| "eval_samples_per_second": 752.905, | |
| "eval_steps_per_second": 23.545, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "grad_norm": 0.09443788975477219, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 10.3177, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.28708133971291866, | |
| "grad_norm": 0.09167303889989853, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 10.3162, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3014354066985646, | |
| "grad_norm": 0.09214193373918533, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 10.3162, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.3014354066985646, | |
| "eval_loss": 10.314752578735352, | |
| "eval_runtime": 1.8438, | |
| "eval_samples_per_second": 763.116, | |
| "eval_steps_per_second": 23.864, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 0.0888906717300415, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 10.3144, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.33014354066985646, | |
| "grad_norm": 0.08958574384450912, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 10.3132, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.3444976076555024, | |
| "grad_norm": 0.08907806128263474, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 10.313, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3444976076555024, | |
| "eval_loss": 10.312495231628418, | |
| "eval_runtime": 1.8465, | |
| "eval_samples_per_second": 761.964, | |
| "eval_steps_per_second": 23.828, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3588516746411483, | |
| "grad_norm": 0.08638429641723633, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 10.3127, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.37320574162679426, | |
| "grad_norm": 0.08639028668403625, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 10.313, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.3875598086124402, | |
| "grad_norm": 0.08610237389802933, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 10.3116, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.3875598086124402, | |
| "eval_loss": 10.311205863952637, | |
| "eval_runtime": 1.8411, | |
| "eval_samples_per_second": 764.228, | |
| "eval_steps_per_second": 23.899, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4019138755980861, | |
| "grad_norm": 0.08259936422109604, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 10.3114, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.41626794258373206, | |
| "grad_norm": 0.08639667928218842, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 10.3115, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.430622009569378, | |
| "grad_norm": 0.08296097069978714, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 10.3116, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.430622009569378, | |
| "eval_loss": 10.310672760009766, | |
| "eval_runtime": 1.8404, | |
| "eval_samples_per_second": 764.503, | |
| "eval_steps_per_second": 23.908, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4449760765550239, | |
| "grad_norm": 0.07877939194440842, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 10.3123, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.45933014354066987, | |
| "grad_norm": 0.08165387064218521, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 10.3103, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "grad_norm": 0.08330517262220383, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 10.3102, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "eval_loss": 10.310574531555176, | |
| "eval_runtime": 1.8448, | |
| "eval_samples_per_second": 762.695, | |
| "eval_steps_per_second": 23.851, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 88348123201536.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |