| { | |
| "best_metric": 0.7025796661608499, | |
| "best_model_checkpoint": "./results/checkpoint-4434", | |
| "epoch": 15.0, | |
| "global_step": 22170, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 2e-06, | |
| "loss": 0.4736, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8308534531162268, | |
| "eval_f1": 0.682141915061989, | |
| "eval_loss": 0.37968066334724426, | |
| "eval_precision": 0.6996753246753247, | |
| "eval_recall": 0.6654657745753989, | |
| "eval_runtime": 2.7102, | |
| "eval_samples_per_second": 68.261, | |
| "eval_steps_per_second": 68.261, | |
| "step": 1478 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 1.9055266887104394e-06, | |
| "loss": 0.3656, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.826221224031443, | |
| "eval_f1": 0.6371629542790153, | |
| "eval_loss": 0.37435096502304077, | |
| "eval_precision": 0.7399591558883595, | |
| "eval_recall": 0.5594441585177561, | |
| "eval_runtime": 2.6167, | |
| "eval_samples_per_second": 70.701, | |
| "eval_steps_per_second": 70.701, | |
| "step": 2956 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.8110533774208785e-06, | |
| "loss": 0.3599, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 1.716580066131318e-06, | |
| "loss": 0.3458, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8349241998877035, | |
| "eval_f1": 0.7025796661608499, | |
| "eval_loss": 0.36284855008125305, | |
| "eval_precision": 0.6907011437095972, | |
| "eval_recall": 0.7148739063304169, | |
| "eval_runtime": 2.6495, | |
| "eval_samples_per_second": 69.825, | |
| "eval_steps_per_second": 69.825, | |
| "step": 4434 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 1.622106754841757e-06, | |
| "loss": 0.337, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8281864121280179, | |
| "eval_f1": 0.651678998292544, | |
| "eval_loss": 0.37390556931495667, | |
| "eval_precision": 0.728835136855506, | |
| "eval_recall": 0.5892949047864128, | |
| "eval_runtime": 2.6125, | |
| "eval_samples_per_second": 70.813, | |
| "eval_steps_per_second": 70.813, | |
| "step": 5912 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 1.5276334435521965e-06, | |
| "loss": 0.3239, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 1.4331601322626356e-06, | |
| "loss": 0.3015, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8357664233576643, | |
| "eval_f1": 0.6978305785123967, | |
| "eval_loss": 0.36176469922065735, | |
| "eval_precision": 0.7003628823224468, | |
| "eval_recall": 0.6953165208440556, | |
| "eval_runtime": 2.8697, | |
| "eval_samples_per_second": 64.467, | |
| "eval_steps_per_second": 64.467, | |
| "step": 7390 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 1.338686820973075e-06, | |
| "loss": 0.2884, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.833941605839416, | |
| "eval_f1": 0.6902330452998167, | |
| "eval_loss": 0.37303170561790466, | |
| "eval_precision": 0.7025586353944563, | |
| "eval_recall": 0.6783324755532681, | |
| "eval_runtime": 2.7072, | |
| "eval_samples_per_second": 68.337, | |
| "eval_steps_per_second": 68.337, | |
| "step": 8868 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 1.2442135096835144e-06, | |
| "loss": 0.264, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 1.1497401983939536e-06, | |
| "loss": 0.2324, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8322571588994947, | |
| "eval_f1": 0.6844467916556641, | |
| "eval_loss": 0.3992096483707428, | |
| "eval_precision": 0.702819956616052, | |
| "eval_recall": 0.6670097786927431, | |
| "eval_runtime": 2.6272, | |
| "eval_samples_per_second": 70.416, | |
| "eval_steps_per_second": 70.416, | |
| "step": 10346 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 1.0552668871043931e-06, | |
| "loss": 0.1965, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8250982594048287, | |
| "eval_f1": 0.6525376464026771, | |
| "eval_loss": 0.44216522574424744, | |
| "eval_precision": 0.7121119902617163, | |
| "eval_recall": 0.602161605764282, | |
| "eval_runtime": 2.622, | |
| "eval_samples_per_second": 70.557, | |
| "eval_steps_per_second": 70.557, | |
| "step": 11824 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 9.607935758148322e-07, | |
| "loss": 0.182, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 8.663202645252716e-07, | |
| "loss": 0.1405, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.823694553621561, | |
| "eval_f1": 0.6694736842105264, | |
| "eval_loss": 0.49459609389305115, | |
| "eval_precision": 0.6849757673667205, | |
| "eval_recall": 0.6546577457539887, | |
| "eval_runtime": 2.6563, | |
| "eval_samples_per_second": 69.645, | |
| "eval_steps_per_second": 69.645, | |
| "step": 13302 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 7.718469532357109e-07, | |
| "loss": 0.1149, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8238349241998877, | |
| "eval_f1": 0.6766297346044834, | |
| "eval_loss": 0.5486593842506409, | |
| "eval_precision": 0.6775025799793601, | |
| "eval_recall": 0.6757591353576943, | |
| "eval_runtime": 2.6698, | |
| "eval_samples_per_second": 69.293, | |
| "eval_steps_per_second": 69.293, | |
| "step": 14780 | |
| }, | |
| { | |
| "epoch": 10.15, | |
| "learning_rate": 6.773736419461502e-07, | |
| "loss": 0.1013, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 10.83, | |
| "learning_rate": 5.829003306565895e-07, | |
| "loss": 0.0808, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8252386299831556, | |
| "eval_f1": 0.6767073487405868, | |
| "eval_loss": 0.6354735493659973, | |
| "eval_precision": 0.6829140461215933, | |
| "eval_recall": 0.6706124549665465, | |
| "eval_runtime": 2.7422, | |
| "eval_samples_per_second": 67.465, | |
| "eval_steps_per_second": 67.465, | |
| "step": 16258 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "learning_rate": 4.884270193670288e-07, | |
| "loss": 0.0627, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8260808534531162, | |
| "eval_f1": 0.6648634027589938, | |
| "eval_loss": 0.6998042464256287, | |
| "eval_precision": 0.7006841505131128, | |
| "eval_recall": 0.6325270200720535, | |
| "eval_runtime": 2.7874, | |
| "eval_samples_per_second": 66.371, | |
| "eval_steps_per_second": 66.371, | |
| "step": 17736 | |
| }, | |
| { | |
| "epoch": 12.18, | |
| "learning_rate": 3.939537080774681e-07, | |
| "loss": 0.0496, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "learning_rate": 2.9948039678790744e-07, | |
| "loss": 0.0399, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8258001122964627, | |
| "eval_f1": 0.6703851261620186, | |
| "eval_loss": 0.7429465651512146, | |
| "eval_precision": 0.6926454445664105, | |
| "eval_recall": 0.649511065362841, | |
| "eval_runtime": 2.7701, | |
| "eval_samples_per_second": 66.784, | |
| "eval_steps_per_second": 66.784, | |
| "step": 19214 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "learning_rate": 2.0500708549834672e-07, | |
| "loss": 0.032, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8218697361033127, | |
| "eval_f1": 0.6556309362279512, | |
| "eval_loss": 0.7868731617927551, | |
| "eval_precision": 0.6934557979334098, | |
| "eval_recall": 0.6217189912506433, | |
| "eval_runtime": 2.7709, | |
| "eval_samples_per_second": 66.766, | |
| "eval_steps_per_second": 66.766, | |
| "step": 20692 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 1.1053377420878601e-07, | |
| "loss": 0.0274, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 14.88, | |
| "learning_rate": 1.606046291922532e-08, | |
| "loss": 0.0233, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8235541830432341, | |
| "eval_f1": 0.6621875839828004, | |
| "eval_loss": 0.7936236262321472, | |
| "eval_precision": 0.6929133858267716, | |
| "eval_recall": 0.6340710241893979, | |
| "eval_runtime": 2.7604, | |
| "eval_samples_per_second": 67.019, | |
| "eval_steps_per_second": 67.019, | |
| "step": 22170 | |
| } | |
| ], | |
| "max_steps": 22170, | |
| "num_train_epochs": 15, | |
| "total_flos": 7912225034580240.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |