| { | |
| "best_metric": 0.8169879527109274, | |
| "best_model_checkpoint": "/home/bram/shares/predict/trained/dutch/hebban-reviews/xlm-roberta-base/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": 4.794166666666667e-05, | |
| "loss": 0.913, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.7665803747534516, | |
| "eval_f1": 0.7649545454029377, | |
| "eval_loss": 0.7253644466400146, | |
| "eval_precision": 0.7641565688780922, | |
| "eval_recall": 0.7665803747534516, | |
| "eval_runtime": 24.3431, | |
| "eval_samples_per_second": 666.472, | |
| "eval_steps_per_second": 27.77, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.5858333333333334e-05, | |
| "loss": 0.7582, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.6934787968441815, | |
| "eval_f1": 0.7165538695299023, | |
| "eval_loss": 0.7026467323303223, | |
| "eval_precision": 0.7706797791371907, | |
| "eval_recall": 0.6934787968441815, | |
| "eval_runtime": 24.1169, | |
| "eval_samples_per_second": 672.723, | |
| "eval_steps_per_second": 28.03, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.3775e-05, | |
| "loss": 0.6847, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.7233727810650887, | |
| "eval_f1": 0.7442026185547411, | |
| "eval_loss": 0.6611877083778381, | |
| "eval_precision": 0.8038626962859567, | |
| "eval_recall": 0.7233727810650887, | |
| "eval_runtime": 24.1771, | |
| "eval_samples_per_second": 671.047, | |
| "eval_steps_per_second": 27.96, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.1691666666666666e-05, | |
| "loss": 0.6532, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_accuracy": 0.8156434911242604, | |
| "eval_f1": 0.8093687626348003, | |
| "eval_loss": 0.6557860374450684, | |
| "eval_precision": 0.8059505966835585, | |
| "eval_recall": 0.8156434911242604, | |
| "eval_runtime": 24.3135, | |
| "eval_samples_per_second": 667.283, | |
| "eval_steps_per_second": 27.803, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.960833333333334e-05, | |
| "loss": 0.6281, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.7695389546351085, | |
| "eval_f1": 0.7795959943899675, | |
| "eval_loss": 0.616236686706543, | |
| "eval_precision": 0.7973850176423627, | |
| "eval_recall": 0.7695389546351085, | |
| "eval_runtime": 24.3023, | |
| "eval_samples_per_second": 667.59, | |
| "eval_steps_per_second": 27.816, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.7525e-05, | |
| "loss": 0.6281, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.8015902366863905, | |
| "eval_f1": 0.803170220169187, | |
| "eval_loss": 0.5991469621658325, | |
| "eval_precision": 0.806066258666526, | |
| "eval_recall": 0.8015902366863905, | |
| "eval_runtime": 24.3429, | |
| "eval_samples_per_second": 666.477, | |
| "eval_steps_per_second": 27.77, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.544583333333333e-05, | |
| "loss": 0.5668, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.7874753451676528, | |
| "eval_f1": 0.7971578305557919, | |
| "eval_loss": 0.5982191562652588, | |
| "eval_precision": 0.8156521173286116, | |
| "eval_recall": 0.7874753451676528, | |
| "eval_runtime": 24.3367, | |
| "eval_samples_per_second": 666.646, | |
| "eval_steps_per_second": 27.777, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.3362500000000005e-05, | |
| "loss": 0.567, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_accuracy": 0.8007889546351085, | |
| "eval_f1": 0.8041408078539543, | |
| "eval_loss": 0.6023094654083252, | |
| "eval_precision": 0.8088979038333125, | |
| "eval_recall": 0.8007889546351085, | |
| "eval_runtime": 24.2641, | |
| "eval_samples_per_second": 668.643, | |
| "eval_steps_per_second": 27.86, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 3.127916666666667e-05, | |
| "loss": 0.5704, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_accuracy": 0.7429117357001972, | |
| "eval_f1": 0.7619810076683907, | |
| "eval_loss": 0.6065093278884888, | |
| "eval_precision": 0.8107750610152921, | |
| "eval_recall": 0.7429117357001972, | |
| "eval_runtime": 24.3353, | |
| "eval_samples_per_second": 666.686, | |
| "eval_steps_per_second": 27.779, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.9195833333333333e-05, | |
| "loss": 0.5596, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_accuracy": 0.8072608481262328, | |
| "eval_f1": 0.8104579115041036, | |
| "eval_loss": 0.5900022983551025, | |
| "eval_precision": 0.8144869336926288, | |
| "eval_recall": 0.8072608481262328, | |
| "eval_runtime": 24.3429, | |
| "eval_samples_per_second": 666.477, | |
| "eval_steps_per_second": 27.77, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 2.7116666666666667e-05, | |
| "loss": 0.5495, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_accuracy": 0.810034516765286, | |
| "eval_f1": 0.8141305380075001, | |
| "eval_loss": 0.613106906414032, | |
| "eval_precision": 0.8219044900382881, | |
| "eval_recall": 0.810034516765286, | |
| "eval_runtime": 24.3418, | |
| "eval_samples_per_second": 666.509, | |
| "eval_steps_per_second": 27.771, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.5033333333333336e-05, | |
| "loss": 0.5449, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.8124383629191322, | |
| "eval_f1": 0.8140798132169556, | |
| "eval_loss": 0.6060279011726379, | |
| "eval_precision": 0.816286158402022, | |
| "eval_recall": 0.8124383629191322, | |
| "eval_runtime": 24.332, | |
| "eval_samples_per_second": 666.777, | |
| "eval_steps_per_second": 27.782, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.2950000000000002e-05, | |
| "loss": 0.4898, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.7848865877712031, | |
| "eval_f1": 0.7964804879952159, | |
| "eval_loss": 0.6215521693229675, | |
| "eval_precision": 0.820125366434727, | |
| "eval_recall": 0.7848865877712031, | |
| "eval_runtime": 24.3196, | |
| "eval_samples_per_second": 667.117, | |
| "eval_steps_per_second": 27.797, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 2.0866666666666668e-05, | |
| "loss": 0.4837, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_accuracy": 0.7318786982248521, | |
| "eval_f1": 0.7528823670992008, | |
| "eval_loss": 0.6411539912223816, | |
| "eval_precision": 0.8100101192694165, | |
| "eval_recall": 0.7318786982248521, | |
| "eval_runtime": 24.3327, | |
| "eval_samples_per_second": 666.758, | |
| "eval_steps_per_second": 27.782, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.87875e-05, | |
| "loss": 0.4671, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_accuracy": 0.803870808678501, | |
| "eval_f1": 0.810953536758238, | |
| "eval_loss": 0.6316830515861511, | |
| "eval_precision": 0.8241951591967538, | |
| "eval_recall": 0.803870808678501, | |
| "eval_runtime": 24.3331, | |
| "eval_samples_per_second": 666.747, | |
| "eval_steps_per_second": 27.781, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 1.670416666666667e-05, | |
| "loss": 0.4791, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_accuracy": 0.8032544378698225, | |
| "eval_f1": 0.8091355876498971, | |
| "eval_loss": 0.5908682942390442, | |
| "eval_precision": 0.8179762946015251, | |
| "eval_recall": 0.8032544378698225, | |
| "eval_runtime": 24.3361, | |
| "eval_samples_per_second": 666.664, | |
| "eval_steps_per_second": 27.778, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.4620833333333334e-05, | |
| "loss": 0.4739, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_accuracy": 0.8067061143984221, | |
| "eval_f1": 0.8120892015198095, | |
| "eval_loss": 0.6165759563446045, | |
| "eval_precision": 0.8199753861011193, | |
| "eval_recall": 0.8067061143984221, | |
| "eval_runtime": 24.1747, | |
| "eval_samples_per_second": 671.116, | |
| "eval_steps_per_second": 27.963, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 1.25375e-05, | |
| "loss": 0.4587, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.8041173570019724, | |
| "eval_f1": 0.8104680056938384, | |
| "eval_loss": 0.5887444019317627, | |
| "eval_precision": 0.820271897699825, | |
| "eval_recall": 0.8041173570019724, | |
| "eval_runtime": 24.1945, | |
| "eval_samples_per_second": 670.564, | |
| "eval_steps_per_second": 27.94, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 1.0454166666666667e-05, | |
| "loss": 0.4147, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.780448717948718, | |
| "eval_f1": 0.7927207824065717, | |
| "eval_loss": 0.6190515160560608, | |
| "eval_precision": 0.8178157106781372, | |
| "eval_recall": 0.780448717948718, | |
| "eval_runtime": 24.1866, | |
| "eval_samples_per_second": 670.784, | |
| "eval_steps_per_second": 27.949, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 8.370833333333333e-06, | |
| "loss": 0.3861, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "eval_accuracy": 0.7917899408284024, | |
| "eval_f1": 0.8013705698417323, | |
| "eval_loss": 0.6606641411781311, | |
| "eval_precision": 0.8190127006775076, | |
| "eval_recall": 0.7917899408284024, | |
| "eval_runtime": 24.3416, | |
| "eval_samples_per_second": 666.514, | |
| "eval_steps_per_second": 27.771, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 6.2875e-06, | |
| "loss": 0.3897, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_accuracy": 0.788646449704142, | |
| "eval_f1": 0.7987527519476123, | |
| "eval_loss": 0.6613931059837341, | |
| "eval_precision": 0.8177063830689373, | |
| "eval_recall": 0.788646449704142, | |
| "eval_runtime": 24.342, | |
| "eval_samples_per_second": 666.503, | |
| "eval_steps_per_second": 27.771, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 4.204166666666667e-06, | |
| "loss": 0.3877, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_accuracy": 0.8135478303747534, | |
| "eval_f1": 0.8169879527109274, | |
| "eval_loss": 0.6640126705169678, | |
| "eval_precision": 0.8215474627622835, | |
| "eval_recall": 0.8135478303747534, | |
| "eval_runtime": 24.3344, | |
| "eval_samples_per_second": 666.71, | |
| "eval_steps_per_second": 27.78, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 2.1208333333333335e-06, | |
| "loss": 0.3795, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_accuracy": 0.8053500986193294, | |
| "eval_f1": 0.8113411583628731, | |
| "eval_loss": 0.6599082350730896, | |
| "eval_precision": 0.8205901740056264, | |
| "eval_recall": 0.8053500986193294, | |
| "eval_runtime": 24.3271, | |
| "eval_samples_per_second": 666.91, | |
| "eval_steps_per_second": 27.788, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 3.7500000000000005e-08, | |
| "loss": 0.3863, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "eval_accuracy": 0.8009122287968442, | |
| "eval_f1": 0.8081790831347971, | |
| "eval_loss": 0.6572125554084778, | |
| "eval_precision": 0.8200806840462704, | |
| "eval_recall": 0.8009122287968442, | |
| "eval_runtime": 24.3366, | |
| "eval_samples_per_second": 666.651, | |
| "eval_steps_per_second": 27.777, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "step": 12000, | |
| "total_flos": 1.4962346001065574e+17, | |
| "train_loss": 0.5341533139546712, | |
| "train_runtime": 3936.6668, | |
| "train_samples_per_second": 146.317, | |
| "train_steps_per_second": 3.048 | |
| } | |
| ], | |
| "max_steps": 12000, | |
| "num_train_epochs": 4, | |
| "total_flos": 1.4962346001065574e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |