| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.48959608323133413, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004895960832313341, | |
| "eval_loss": 3.1936099529266357, | |
| "eval_runtime": 9.9463, | |
| "eval_samples_per_second": 34.586, | |
| "eval_steps_per_second": 4.323, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014687882496940025, | |
| "grad_norm": 1.0404754877090454, | |
| "learning_rate": 3e-05, | |
| "loss": 12.5665, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02937576499388005, | |
| "grad_norm": 0.9915498495101929, | |
| "learning_rate": 6e-05, | |
| "loss": 12.4935, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.044063647490820076, | |
| "grad_norm": 1.0037130117416382, | |
| "learning_rate": 9e-05, | |
| "loss": 12.1925, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.044063647490820076, | |
| "eval_loss": 3.181182861328125, | |
| "eval_runtime": 9.9271, | |
| "eval_samples_per_second": 34.653, | |
| "eval_steps_per_second": 4.332, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0587515299877601, | |
| "grad_norm": 1.1385695934295654, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 12.5931, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.07343941248470012, | |
| "grad_norm": 1.1365324258804321, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 12.6559, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08812729498164015, | |
| "grad_norm": 1.352217674255371, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 12.2752, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08812729498164015, | |
| "eval_loss": 3.085514545440674, | |
| "eval_runtime": 9.9359, | |
| "eval_samples_per_second": 34.622, | |
| "eval_steps_per_second": 4.328, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10281517747858017, | |
| "grad_norm": 1.6765836477279663, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 12.1162, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1175030599755202, | |
| "grad_norm": 1.3779009580612183, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 12.0488, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.13219094247246022, | |
| "grad_norm": 1.1408171653747559, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 12.1308, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.13219094247246022, | |
| "eval_loss": 2.9765007495880127, | |
| "eval_runtime": 9.9142, | |
| "eval_samples_per_second": 34.698, | |
| "eval_steps_per_second": 4.337, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.14687882496940025, | |
| "grad_norm": 1.1627850532531738, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 11.5805, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16156670746634028, | |
| "grad_norm": 1.1915265321731567, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 11.6809, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1762545899632803, | |
| "grad_norm": 1.0108513832092285, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 11.317, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1762545899632803, | |
| "eval_loss": 2.90360164642334, | |
| "eval_runtime": 9.9067, | |
| "eval_samples_per_second": 34.724, | |
| "eval_steps_per_second": 4.34, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1909424724602203, | |
| "grad_norm": 1.106205940246582, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 11.3996, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.20563035495716034, | |
| "grad_norm": 1.2400563955307007, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 11.3239, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.22031823745410037, | |
| "grad_norm": 1.2355320453643799, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 11.3475, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.22031823745410037, | |
| "eval_loss": 2.8525941371917725, | |
| "eval_runtime": 9.9142, | |
| "eval_samples_per_second": 34.698, | |
| "eval_steps_per_second": 4.337, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2350061199510404, | |
| "grad_norm": 1.0597670078277588, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 11.768, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.24969400244798043, | |
| "grad_norm": 1.2998725175857544, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 11.5956, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.26438188494492043, | |
| "grad_norm": 1.1917494535446167, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 10.8064, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.26438188494492043, | |
| "eval_loss": 2.8190228939056396, | |
| "eval_runtime": 9.9101, | |
| "eval_samples_per_second": 34.712, | |
| "eval_steps_per_second": 4.339, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.27906976744186046, | |
| "grad_norm": 1.0400537252426147, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 11.0586, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.2937576499388005, | |
| "grad_norm": 1.1486033201217651, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 11.2156, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3084455324357405, | |
| "grad_norm": 1.1027334928512573, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 10.92, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.3084455324357405, | |
| "eval_loss": 2.7967541217803955, | |
| "eval_runtime": 9.9168, | |
| "eval_samples_per_second": 34.689, | |
| "eval_steps_per_second": 4.336, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.32313341493268055, | |
| "grad_norm": 1.0224430561065674, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 11.4176, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.3378212974296206, | |
| "grad_norm": 1.2892149686813354, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 11.1174, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.3525091799265606, | |
| "grad_norm": 1.0721704959869385, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 10.7613, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3525091799265606, | |
| "eval_loss": 2.78279185295105, | |
| "eval_runtime": 9.9215, | |
| "eval_samples_per_second": 34.672, | |
| "eval_steps_per_second": 4.334, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3671970624235006, | |
| "grad_norm": 1.1785831451416016, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 11.1049, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3818849449204406, | |
| "grad_norm": 1.1622775793075562, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 10.7311, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.39657282741738065, | |
| "grad_norm": 1.2672510147094727, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 11.153, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.39657282741738065, | |
| "eval_loss": 2.775190830230713, | |
| "eval_runtime": 9.9097, | |
| "eval_samples_per_second": 34.713, | |
| "eval_steps_per_second": 4.339, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4112607099143207, | |
| "grad_norm": 1.121291995048523, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 11.0814, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.4259485924112607, | |
| "grad_norm": 1.2836380004882812, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 11.0664, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.44063647490820074, | |
| "grad_norm": 1.186371922492981, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 10.7842, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.44063647490820074, | |
| "eval_loss": 2.772207498550415, | |
| "eval_runtime": 9.9076, | |
| "eval_samples_per_second": 34.721, | |
| "eval_steps_per_second": 4.34, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.45532435740514077, | |
| "grad_norm": 1.218754768371582, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 10.7893, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.4700122399020808, | |
| "grad_norm": 1.0495154857635498, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 10.9183, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.4847001223990208, | |
| "grad_norm": 0.9865614771842957, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 11.1974, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.4847001223990208, | |
| "eval_loss": 2.7715492248535156, | |
| "eval_runtime": 9.9162, | |
| "eval_samples_per_second": 34.691, | |
| "eval_steps_per_second": 4.336, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3447818551296000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |