| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 31.998864926220204, | |
| "global_step": 7040, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8224893386183709, | |
| "eval_loss": 1.1622651815414429, | |
| "eval_runtime": 5.9816, | |
| "eval_samples_per_second": 15.213, | |
| "eval_steps_per_second": 7.69, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8538665635439829, | |
| "eval_loss": 0.9566460251808167, | |
| "eval_runtime": 5.9768, | |
| "eval_samples_per_second": 15.226, | |
| "eval_steps_per_second": 7.696, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 4.6448863636363636e-05, | |
| "loss": 1.1942, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8709140322043548, | |
| "eval_loss": 0.8456286191940308, | |
| "eval_runtime": 5.9871, | |
| "eval_samples_per_second": 15.199, | |
| "eval_steps_per_second": 7.683, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8801198801198801, | |
| "eval_loss": 0.7718582153320312, | |
| "eval_runtime": 5.6728, | |
| "eval_samples_per_second": 16.041, | |
| "eval_steps_per_second": 8.109, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 4.289772727272727e-05, | |
| "loss": 0.7805, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8871880807364678, | |
| "eval_loss": 0.7224407196044922, | |
| "eval_runtime": 5.988, | |
| "eval_samples_per_second": 15.197, | |
| "eval_steps_per_second": 7.682, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.892816860558796, | |
| "eval_loss": 0.6894707679748535, | |
| "eval_runtime": 5.9958, | |
| "eval_samples_per_second": 15.177, | |
| "eval_steps_per_second": 7.672, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 3.934659090909091e-05, | |
| "loss": 0.6257, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8972103165651553, | |
| "eval_loss": 0.6574080586433411, | |
| "eval_runtime": 5.9909, | |
| "eval_samples_per_second": 15.19, | |
| "eval_steps_per_second": 7.678, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9014426433781273, | |
| "eval_loss": 0.6289474368095398, | |
| "eval_runtime": 5.6911, | |
| "eval_samples_per_second": 15.99, | |
| "eval_steps_per_second": 8.083, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9045470658373884, | |
| "eval_loss": 0.6054205298423767, | |
| "eval_runtime": 5.6786, | |
| "eval_samples_per_second": 16.025, | |
| "eval_steps_per_second": 8.101, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 3.579545454545455e-05, | |
| "loss": 0.5385, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9076622302428754, | |
| "eval_loss": 0.5881273746490479, | |
| "eval_runtime": 6.0014, | |
| "eval_samples_per_second": 15.163, | |
| "eval_steps_per_second": 7.665, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9101865876059424, | |
| "eval_loss": 0.5709272623062134, | |
| "eval_runtime": 5.9886, | |
| "eval_samples_per_second": 15.195, | |
| "eval_steps_per_second": 7.681, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "learning_rate": 3.2244318181818185e-05, | |
| "loss": 0.4778, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9120986540341379, | |
| "eval_loss": 0.5591339468955994, | |
| "eval_runtime": 5.9874, | |
| "eval_samples_per_second": 15.199, | |
| "eval_steps_per_second": 7.683, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9142900110642046, | |
| "eval_loss": 0.5496613383293152, | |
| "eval_runtime": 5.9826, | |
| "eval_samples_per_second": 15.211, | |
| "eval_steps_per_second": 7.689, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "learning_rate": 2.870028409090909e-05, | |
| "loss": 0.427, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9161053999763678, | |
| "eval_loss": 0.5385328531265259, | |
| "eval_runtime": 5.9885, | |
| "eval_samples_per_second": 15.196, | |
| "eval_steps_per_second": 7.681, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9179637566734341, | |
| "eval_loss": 0.5258467793464661, | |
| "eval_runtime": 5.9812, | |
| "eval_samples_per_second": 15.214, | |
| "eval_steps_per_second": 7.691, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 2.5149147727272725e-05, | |
| "loss": 0.394, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9195428227686292, | |
| "eval_loss": 0.5170450806617737, | |
| "eval_runtime": 5.6775, | |
| "eval_samples_per_second": 16.028, | |
| "eval_steps_per_second": 8.102, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9211648566487276, | |
| "eval_loss": 0.5156892538070679, | |
| "eval_runtime": 5.9945, | |
| "eval_samples_per_second": 15.18, | |
| "eval_steps_per_second": 7.674, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9220671801316963, | |
| "eval_loss": 0.5037761926651001, | |
| "eval_runtime": 5.687, | |
| "eval_samples_per_second": 16.001, | |
| "eval_steps_per_second": 8.089, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 2.1598011363636363e-05, | |
| "loss": 0.363, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9233884395174717, | |
| "eval_loss": 0.49766021966934204, | |
| "eval_runtime": 5.6849, | |
| "eval_samples_per_second": 16.007, | |
| "eval_steps_per_second": 8.092, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9236462462268914, | |
| "eval_loss": 0.4975946843624115, | |
| "eval_runtime": 5.9923, | |
| "eval_samples_per_second": 15.186, | |
| "eval_steps_per_second": 7.677, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 20.45, | |
| "learning_rate": 1.8046875000000003e-05, | |
| "loss": 0.3392, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.9246882150107957, | |
| "eval_loss": 0.49241966009140015, | |
| "eval_runtime": 5.9923, | |
| "eval_samples_per_second": 15.186, | |
| "eval_steps_per_second": 7.677, | |
| "step": 4620 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.9255046029239578, | |
| "eval_loss": 0.4887617826461792, | |
| "eval_runtime": 5.69, | |
| "eval_samples_per_second": 15.993, | |
| "eval_steps_per_second": 8.084, | |
| "step": 4840 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "learning_rate": 1.4495738636363637e-05, | |
| "loss": 0.33, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.9262028294286359, | |
| "eval_loss": 0.4889785051345825, | |
| "eval_runtime": 6.0024, | |
| "eval_samples_per_second": 15.161, | |
| "eval_steps_per_second": 7.664, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9267936364710558, | |
| "eval_loss": 0.4856303334236145, | |
| "eval_runtime": 5.9867, | |
| "eval_samples_per_second": 15.2, | |
| "eval_steps_per_second": 7.684, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 1.0951704545454545e-05, | |
| "loss": 0.3058, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.9275348307606373, | |
| "eval_loss": 0.4802783131599426, | |
| "eval_runtime": 5.6869, | |
| "eval_samples_per_second": 16.002, | |
| "eval_steps_per_second": 8.089, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.9277389277389277, | |
| "eval_loss": 0.47845765948295593, | |
| "eval_runtime": 5.9773, | |
| "eval_samples_per_second": 15.224, | |
| "eval_steps_per_second": 7.696, | |
| "step": 5720 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.9280611861257022, | |
| "eval_loss": 0.4813348948955536, | |
| "eval_runtime": 5.6869, | |
| "eval_samples_per_second": 16.002, | |
| "eval_steps_per_second": 8.089, | |
| "step": 5940 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 7.407670454545455e-06, | |
| "loss": 0.2973, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9281793475341863, | |
| "eval_loss": 0.4798637628555298, | |
| "eval_runtime": 5.6812, | |
| "eval_samples_per_second": 16.018, | |
| "eval_steps_per_second": 8.097, | |
| "step": 6160 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.9285230898134124, | |
| "eval_loss": 0.47730037569999695, | |
| "eval_runtime": 5.9912, | |
| "eval_samples_per_second": 15.189, | |
| "eval_steps_per_second": 7.678, | |
| "step": 6380 | |
| }, | |
| { | |
| "epoch": 29.54, | |
| "learning_rate": 3.856534090909091e-06, | |
| "loss": 0.2931, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.9285982834369931, | |
| "eval_loss": 0.4778377413749695, | |
| "eval_runtime": 5.9957, | |
| "eval_samples_per_second": 15.177, | |
| "eval_steps_per_second": 7.672, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.9289527676624451, | |
| "eval_loss": 0.4756244122982025, | |
| "eval_runtime": 5.6812, | |
| "eval_samples_per_second": 16.018, | |
| "eval_steps_per_second": 8.097, | |
| "step": 6820 | |
| }, | |
| { | |
| "epoch": 31.82, | |
| "learning_rate": 3.053977272727273e-07, | |
| "loss": 0.2879, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9289527676624451, | |
| "eval_loss": 0.47753414511680603, | |
| "eval_runtime": 5.6774, | |
| "eval_samples_per_second": 16.029, | |
| "eval_steps_per_second": 8.102, | |
| "step": 7040 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "step": 7040, | |
| "total_flos": 2.4539422830415053e+17, | |
| "train_loss": 0.47427067851478405, | |
| "train_runtime": 10360.152, | |
| "train_samples_per_second": 5.442, | |
| "train_steps_per_second": 0.68 | |
| } | |
| ], | |
| "max_steps": 7040, | |
| "num_train_epochs": 32, | |
| "total_flos": 2.4539422830415053e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |