| { | |
| "best_metric": 0.7886478688599108, | |
| "best_model_checkpoint": "/home/bram/shares/predict/trained/dutch/hebban-reviews/bert-base-multilingual-cased/checkpoint-11000", | |
| "epoch": 3.9447731755424065, | |
| "global_step": 12000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.793333333333334e-05, | |
| "loss": 0.9625, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.6706114398422091, | |
| "eval_f1": 0.6931944548479266, | |
| "eval_loss": 0.8083919882774353, | |
| "eval_precision": 0.7343487735413422, | |
| "eval_recall": 0.6706114398422091, | |
| "eval_runtime": 24.6562, | |
| "eval_samples_per_second": 658.009, | |
| "eval_steps_per_second": 27.417, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.585e-05, | |
| "loss": 0.8186, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.6870685404339251, | |
| "eval_f1": 0.7088564060365691, | |
| "eval_loss": 0.7797691226005554, | |
| "eval_precision": 0.753903679506183, | |
| "eval_recall": 0.6870685404339251, | |
| "eval_runtime": 24.786, | |
| "eval_samples_per_second": 654.562, | |
| "eval_steps_per_second": 27.273, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.376666666666667e-05, | |
| "loss": 0.7658, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.7040803747534516, | |
| "eval_f1": 0.7250925730375316, | |
| "eval_loss": 0.7284208536148071, | |
| "eval_precision": 0.777831708888876, | |
| "eval_recall": 0.7040803747534516, | |
| "eval_runtime": 24.7754, | |
| "eval_samples_per_second": 654.842, | |
| "eval_steps_per_second": 27.285, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.1683333333333335e-05, | |
| "loss": 0.7318, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_accuracy": 0.7891395463510849, | |
| "eval_f1": 0.7822709378126483, | |
| "eval_loss": 0.7148900032043457, | |
| "eval_precision": 0.7780706309716242, | |
| "eval_recall": 0.7891395463510849, | |
| "eval_runtime": 24.6279, | |
| "eval_samples_per_second": 658.766, | |
| "eval_steps_per_second": 27.449, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.960000000000001e-05, | |
| "loss": 0.6977, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.7249753451676528, | |
| "eval_f1": 0.7429677966966411, | |
| "eval_loss": 0.6928562521934509, | |
| "eval_precision": 0.7807492770659993, | |
| "eval_recall": 0.7249753451676528, | |
| "eval_runtime": 24.6337, | |
| "eval_samples_per_second": 658.61, | |
| "eval_steps_per_second": 27.442, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.7516666666666666e-05, | |
| "loss": 0.7032, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.7484590729783037, | |
| "eval_f1": 0.7594497306385251, | |
| "eval_loss": 0.6671503782272339, | |
| "eval_precision": 0.7788847542194782, | |
| "eval_recall": 0.7484590729783037, | |
| "eval_runtime": 24.5732, | |
| "eval_samples_per_second": 660.231, | |
| "eval_steps_per_second": 27.51, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.543333333333333e-05, | |
| "loss": 0.6227, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.7747781065088757, | |
| "eval_f1": 0.7788314002376725, | |
| "eval_loss": 0.6905977129936218, | |
| "eval_precision": 0.7840673104698108, | |
| "eval_recall": 0.7747781065088757, | |
| "eval_runtime": 24.7721, | |
| "eval_samples_per_second": 654.931, | |
| "eval_steps_per_second": 27.289, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.3350000000000004e-05, | |
| "loss": 0.6309, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_accuracy": 0.7607864891518737, | |
| "eval_f1": 0.7709664128632473, | |
| "eval_loss": 0.6686555743217468, | |
| "eval_precision": 0.7878354043029604, | |
| "eval_recall": 0.7607864891518737, | |
| "eval_runtime": 24.6117, | |
| "eval_samples_per_second": 659.198, | |
| "eval_steps_per_second": 27.467, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 3.126666666666666e-05, | |
| "loss": 0.6256, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_accuracy": 0.6715359960552268, | |
| "eval_f1": 0.7001277240194975, | |
| "eval_loss": 0.6691098809242249, | |
| "eval_precision": 0.7948795089365344, | |
| "eval_recall": 0.6715359960552268, | |
| "eval_runtime": 24.6266, | |
| "eval_samples_per_second": 658.801, | |
| "eval_steps_per_second": 27.45, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.91875e-05, | |
| "loss": 0.6152, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_accuracy": 0.7810034516765286, | |
| "eval_f1": 0.7815583569804468, | |
| "eval_loss": 0.6452860832214355, | |
| "eval_precision": 0.7840264023525906, | |
| "eval_recall": 0.7810034516765286, | |
| "eval_runtime": 24.5688, | |
| "eval_samples_per_second": 660.348, | |
| "eval_steps_per_second": 27.515, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 2.710416666666667e-05, | |
| "loss": 0.6068, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_accuracy": 0.7764423076923077, | |
| "eval_f1": 0.7843276562987542, | |
| "eval_loss": 0.6676753759384155, | |
| "eval_precision": 0.8001617824983851, | |
| "eval_recall": 0.7764423076923077, | |
| "eval_runtime": 24.7224, | |
| "eval_samples_per_second": 656.247, | |
| "eval_steps_per_second": 27.344, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.5020833333333333e-05, | |
| "loss": 0.6016, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.7220784023668639, | |
| "eval_f1": 0.7405665635983271, | |
| "eval_loss": 0.6461244821548462, | |
| "eval_precision": 0.782308061609766, | |
| "eval_recall": 0.7220784023668639, | |
| "eval_runtime": 24.6746, | |
| "eval_samples_per_second": 657.518, | |
| "eval_steps_per_second": 27.397, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.2937500000000002e-05, | |
| "loss": 0.521, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.7281188362919132, | |
| "eval_f1": 0.7466913370432308, | |
| "eval_loss": 0.7071252465248108, | |
| "eval_precision": 0.7878399669333115, | |
| "eval_recall": 0.7281188362919132, | |
| "eval_runtime": 24.7635, | |
| "eval_samples_per_second": 655.158, | |
| "eval_steps_per_second": 27.298, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 2.0854166666666668e-05, | |
| "loss": 0.5137, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_accuracy": 0.7346523668639053, | |
| "eval_f1": 0.752395675119321, | |
| "eval_loss": 0.6949610114097595, | |
| "eval_precision": 0.7902262301474485, | |
| "eval_recall": 0.7346523668639053, | |
| "eval_runtime": 24.761, | |
| "eval_samples_per_second": 655.225, | |
| "eval_steps_per_second": 27.301, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.8770833333333333e-05, | |
| "loss": 0.504, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_accuracy": 0.7453772189349113, | |
| "eval_f1": 0.7618113431887504, | |
| "eval_loss": 0.6824287176132202, | |
| "eval_precision": 0.7979845371909847, | |
| "eval_recall": 0.7453772189349113, | |
| "eval_runtime": 24.7586, | |
| "eval_samples_per_second": 655.287, | |
| "eval_steps_per_second": 27.304, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 1.6691666666666668e-05, | |
| "loss": 0.5049, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_accuracy": 0.7816814595660749, | |
| "eval_f1": 0.7878816281164417, | |
| "eval_loss": 0.6992799043655396, | |
| "eval_precision": 0.7967192296633613, | |
| "eval_recall": 0.7816814595660749, | |
| "eval_runtime": 24.7558, | |
| "eval_samples_per_second": 655.361, | |
| "eval_steps_per_second": 27.307, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.4608333333333335e-05, | |
| "loss": 0.4987, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_accuracy": 0.7680596646942801, | |
| "eval_f1": 0.779595614035326, | |
| "eval_loss": 0.7158842086791992, | |
| "eval_precision": 0.8023493022129313, | |
| "eval_recall": 0.7680596646942801, | |
| "eval_runtime": 24.6159, | |
| "eval_samples_per_second": 659.087, | |
| "eval_steps_per_second": 27.462, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 1.2525000000000001e-05, | |
| "loss": 0.4978, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.7757026627218935, | |
| "eval_f1": 0.7846497142505, | |
| "eval_loss": 0.6755026578903198, | |
| "eval_precision": 0.7992058416315558, | |
| "eval_recall": 0.7757026627218935, | |
| "eval_runtime": 24.6076, | |
| "eval_samples_per_second": 659.309, | |
| "eval_steps_per_second": 27.471, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 1.0441666666666667e-05, | |
| "loss": 0.4236, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.7638067061143984, | |
| "eval_f1": 0.7757104932531658, | |
| "eval_loss": 0.740465521812439, | |
| "eval_precision": 0.7974750781708845, | |
| "eval_recall": 0.7638067061143984, | |
| "eval_runtime": 24.6338, | |
| "eval_samples_per_second": 658.607, | |
| "eval_steps_per_second": 27.442, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 8.358333333333333e-06, | |
| "loss": 0.3987, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "eval_accuracy": 0.7555473372781065, | |
| "eval_f1": 0.7701303707262471, | |
| "eval_loss": 0.7569716572761536, | |
| "eval_precision": 0.80053210602621, | |
| "eval_recall": 0.7555473372781065, | |
| "eval_runtime": 24.6348, | |
| "eval_samples_per_second": 658.58, | |
| "eval_steps_per_second": 27.441, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 6.275e-06, | |
| "loss": 0.3914, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_accuracy": 0.7660256410256411, | |
| "eval_f1": 0.7781937770368635, | |
| "eval_loss": 0.7907547354698181, | |
| "eval_precision": 0.8023937992057022, | |
| "eval_recall": 0.7660256410256411, | |
| "eval_runtime": 24.7673, | |
| "eval_samples_per_second": 655.057, | |
| "eval_steps_per_second": 27.294, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 4.195833333333334e-06, | |
| "loss": 0.4007, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_accuracy": 0.7821745562130178, | |
| "eval_f1": 0.7886478688599108, | |
| "eval_loss": 0.7599502801895142, | |
| "eval_precision": 0.7983440482619892, | |
| "eval_recall": 0.7821745562130178, | |
| "eval_runtime": 24.6844, | |
| "eval_samples_per_second": 657.258, | |
| "eval_steps_per_second": 27.386, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 2.1125e-06, | |
| "loss": 0.3878, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_accuracy": 0.7740384615384616, | |
| "eval_f1": 0.7836077149829459, | |
| "eval_loss": 0.7647958993911743, | |
| "eval_precision": 0.8007153395014647, | |
| "eval_recall": 0.7740384615384616, | |
| "eval_runtime": 24.6189, | |
| "eval_samples_per_second": 659.006, | |
| "eval_steps_per_second": 27.459, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 2.916666666666667e-08, | |
| "loss": 0.3829, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "eval_accuracy": 0.7662721893491125, | |
| "eval_f1": 0.7779778325177736, | |
| "eval_loss": 0.752331018447876, | |
| "eval_precision": 0.7995290654452684, | |
| "eval_recall": 0.7662721893491125, | |
| "eval_runtime": 24.6488, | |
| "eval_samples_per_second": 658.206, | |
| "eval_steps_per_second": 27.425, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "step": 12000, | |
| "total_flos": 1.502407617014661e+17, | |
| "train_loss": 0.5753211771647135, | |
| "train_runtime": 3636.5588, | |
| "train_samples_per_second": 158.391, | |
| "train_steps_per_second": 3.3 | |
| } | |
| ], | |
| "max_steps": 12000, | |
| "num_train_epochs": 4, | |
| "total_flos": 1.502407617014661e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |