| { | |
| "best_metric": 0.9820269110269746, | |
| "best_model_checkpoint": "nreimers/MiniLMv2-L6-H384-distilled-from-BERT-Large-finetuned-toxic/checkpoint-9974", | |
| "epoch": 3.0, | |
| "eval_steps": 1000, | |
| "global_step": 14961, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.0024057738572574e-05, | |
| "loss": 0.613, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.0048115477145148e-05, | |
| "loss": 0.1678, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 3.0072173215717726e-05, | |
| "loss": 0.0761, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.0096230954290296e-05, | |
| "loss": 0.0607, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.9986631611782006e-05, | |
| "loss": 0.0544, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.887259926028252e-05, | |
| "loss": 0.0538, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.775856690878303e-05, | |
| "loss": 0.0518, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.6644534557283545e-05, | |
| "loss": 0.0476, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.553050220578406e-05, | |
| "loss": 0.0468, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.07939296215772629, | |
| "eval_roc_auc_score": 0.9785734350354863, | |
| "eval_runtime": 1456.3659, | |
| "eval_samples_per_second": 43.93, | |
| "eval_steps_per_second": 1.373, | |
| "step": 4987 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.441646985428457e-05, | |
| "loss": 0.0467, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 4.330243750278508e-05, | |
| "loss": 0.0404, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.218840515128559e-05, | |
| "loss": 0.0406, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 4.107437279978611e-05, | |
| "loss": 0.0391, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 3.996034044828662e-05, | |
| "loss": 0.0396, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 3.884630809678713e-05, | |
| "loss": 0.0402, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.7732275745287644e-05, | |
| "loss": 0.0378, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 3.661824339378816e-05, | |
| "loss": 0.0376, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 3.550421104228867e-05, | |
| "loss": 0.0375, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 3.439017869078918e-05, | |
| "loss": 0.036, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.06088978424668312, | |
| "eval_roc_auc_score": 0.9820269110269746, | |
| "eval_runtime": 1456.3643, | |
| "eval_samples_per_second": 43.93, | |
| "eval_steps_per_second": 1.373, | |
| "step": 9974 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 3.3276146339289697e-05, | |
| "loss": 0.0369, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 3.21621139877902e-05, | |
| "loss": 0.0322, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.104808163629072e-05, | |
| "loss": 0.0302, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 2.9934049284791233e-05, | |
| "loss": 0.0301, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 2.8820016933291742e-05, | |
| "loss": 0.0303, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 2.7705984581792256e-05, | |
| "loss": 0.0298, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 2.6591952230292772e-05, | |
| "loss": 0.0306, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 2.5477919878793282e-05, | |
| "loss": 0.0312, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 2.4363887527293795e-05, | |
| "loss": 0.0308, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 2.3249855175794305e-05, | |
| "loss": 0.0301, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.061725061386823654, | |
| "eval_roc_auc_score": 0.9815256646489099, | |
| "eval_runtime": 1457.0586, | |
| "eval_samples_per_second": 43.909, | |
| "eval_steps_per_second": 1.373, | |
| "step": 14961 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 24935, | |
| "num_train_epochs": 5, | |
| "save_steps": 1000, | |
| "total_flos": 1.5876904241682432e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |