| { | |
| "best_metric": 0.826051364529945, | |
| "best_model_checkpoint": "/home/bram/shares/predict/trained/dutch/hebban-reviews/robbert-v2-dutch-base/checkpoint-2500", | |
| "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": 4.7925000000000006e-05, | |
| "loss": 0.7262, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.7467948717948718, | |
| "eval_f1": 0.7588505397999891, | |
| "eval_loss": 0.6503846049308777, | |
| "eval_precision": 0.7826608700318252, | |
| "eval_recall": 0.7467948717948718, | |
| "eval_runtime": 24.2662, | |
| "eval_samples_per_second": 668.584, | |
| "eval_steps_per_second": 27.858, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.5841666666666665e-05, | |
| "loss": 0.6388, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.7707716962524654, | |
| "eval_f1": 0.7828012503260621, | |
| "eval_loss": 0.6046772599220276, | |
| "eval_precision": 0.806817293633085, | |
| "eval_recall": 0.7707716962524654, | |
| "eval_runtime": 24.3206, | |
| "eval_samples_per_second": 667.089, | |
| "eval_steps_per_second": 27.795, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.375833333333334e-05, | |
| "loss": 0.6036, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.7991247534516766, | |
| "eval_f1": 0.8068009793163453, | |
| "eval_loss": 0.5817954540252686, | |
| "eval_precision": 0.8200694446499367, | |
| "eval_recall": 0.7991247534516766, | |
| "eval_runtime": 24.3415, | |
| "eval_samples_per_second": 666.516, | |
| "eval_steps_per_second": 27.771, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.1675e-05, | |
| "loss": 0.5894, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_accuracy": 0.8214373767258383, | |
| "eval_f1": 0.8223555383757819, | |
| "eval_loss": 0.5838125944137573, | |
| "eval_precision": 0.8240754296334577, | |
| "eval_recall": 0.8214373767258383, | |
| "eval_runtime": 24.3264, | |
| "eval_samples_per_second": 666.929, | |
| "eval_steps_per_second": 27.789, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.960000000000001e-05, | |
| "loss": 0.5678, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.8257519723865878, | |
| "eval_f1": 0.826051364529945, | |
| "eval_loss": 0.565325915813446, | |
| "eval_precision": 0.8264063729685338, | |
| "eval_recall": 0.8257519723865878, | |
| "eval_runtime": 24.3378, | |
| "eval_samples_per_second": 666.617, | |
| "eval_steps_per_second": 27.776, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.7516666666666666e-05, | |
| "loss": 0.5821, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.8164447731755424, | |
| "eval_f1": 0.8171577310703354, | |
| "eval_loss": 0.5594866275787354, | |
| "eval_precision": 0.8191229796098939, | |
| "eval_recall": 0.8164447731755424, | |
| "eval_runtime": 24.3358, | |
| "eval_samples_per_second": 666.673, | |
| "eval_steps_per_second": 27.778, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.543333333333333e-05, | |
| "loss": 0.4741, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.7975838264299803, | |
| "eval_f1": 0.8063837559416328, | |
| "eval_loss": 0.5798861384391785, | |
| "eval_precision": 0.8218943463788534, | |
| "eval_recall": 0.7975838264299803, | |
| "eval_runtime": 24.3305, | |
| "eval_samples_per_second": 666.816, | |
| "eval_steps_per_second": 27.784, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.3350000000000004e-05, | |
| "loss": 0.4763, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_accuracy": 0.8000493096646942, | |
| "eval_f1": 0.8080426775793622, | |
| "eval_loss": 0.5934886336326599, | |
| "eval_precision": 0.8221336258804294, | |
| "eval_recall": 0.8000493096646942, | |
| "eval_runtime": 24.3209, | |
| "eval_samples_per_second": 667.08, | |
| "eval_steps_per_second": 27.795, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 3.126666666666666e-05, | |
| "loss": 0.4719, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_accuracy": 0.7595537475345168, | |
| "eval_f1": 0.7769614913537937, | |
| "eval_loss": 0.5927842855453491, | |
| "eval_precision": 0.8215588992651691, | |
| "eval_recall": 0.7595537475345168, | |
| "eval_runtime": 24.3097, | |
| "eval_samples_per_second": 667.388, | |
| "eval_steps_per_second": 27.808, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.91875e-05, | |
| "loss": 0.477, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_accuracy": 0.8143491124260355, | |
| "eval_f1": 0.8178014358117583, | |
| "eval_loss": 0.571167528629303, | |
| "eval_precision": 0.8224283181214769, | |
| "eval_recall": 0.8143491124260355, | |
| "eval_runtime": 24.3019, | |
| "eval_samples_per_second": 667.601, | |
| "eval_steps_per_second": 27.817, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 2.710416666666667e-05, | |
| "loss": 0.4627, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_accuracy": 0.8163831360946746, | |
| "eval_f1": 0.8209409596272614, | |
| "eval_loss": 0.5852541327476501, | |
| "eval_precision": 0.8276030279019199, | |
| "eval_recall": 0.8163831360946746, | |
| "eval_runtime": 24.3288, | |
| "eval_samples_per_second": 666.863, | |
| "eval_steps_per_second": 27.786, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.5020833333333333e-05, | |
| "loss": 0.4669, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.823594674556213, | |
| "eval_f1": 0.8244437619387351, | |
| "eval_loss": 0.5960806608200073, | |
| "eval_precision": 0.8258010274342865, | |
| "eval_recall": 0.823594674556213, | |
| "eval_runtime": 24.228, | |
| "eval_samples_per_second": 669.639, | |
| "eval_steps_per_second": 27.902, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.2937500000000002e-05, | |
| "loss": 0.3604, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.8014669625246549, | |
| "eval_f1": 0.8102801213854947, | |
| "eval_loss": 0.6862322092056274, | |
| "eval_precision": 0.8275335401956307, | |
| "eval_recall": 0.8014669625246549, | |
| "eval_runtime": 24.2792, | |
| "eval_samples_per_second": 668.227, | |
| "eval_steps_per_second": 27.843, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 2.0854166666666668e-05, | |
| "loss": 0.3489, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_accuracy": 0.7792776134122288, | |
| "eval_f1": 0.7927818598604747, | |
| "eval_loss": 0.6760165095329285, | |
| "eval_precision": 0.822269294544987, | |
| "eval_recall": 0.7792776134122288, | |
| "eval_runtime": 24.2892, | |
| "eval_samples_per_second": 667.95, | |
| "eval_steps_per_second": 27.831, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.8775000000000002e-05, | |
| "loss": 0.3394, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_accuracy": 0.8185404339250493, | |
| "eval_f1": 0.8224920113739561, | |
| "eval_loss": 0.6916875839233398, | |
| "eval_precision": 0.8283683902286886, | |
| "eval_recall": 0.8185404339250493, | |
| "eval_runtime": 24.2568, | |
| "eval_samples_per_second": 668.844, | |
| "eval_steps_per_second": 27.868, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 1.6695833333333334e-05, | |
| "loss": 0.3462, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_accuracy": 0.8147805719921104, | |
| "eval_f1": 0.8200537266341018, | |
| "eval_loss": 0.6634742617607117, | |
| "eval_precision": 0.8279502059904518, | |
| "eval_recall": 0.8147805719921104, | |
| "eval_runtime": 24.237, | |
| "eval_samples_per_second": 669.39, | |
| "eval_steps_per_second": 27.891, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.4612500000000001e-05, | |
| "loss": 0.3498, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_accuracy": 0.8173076923076923, | |
| "eval_f1": 0.8219188790015944, | |
| "eval_loss": 0.6741124987602234, | |
| "eval_precision": 0.8291747460835226, | |
| "eval_recall": 0.8173076923076923, | |
| "eval_runtime": 24.3305, | |
| "eval_samples_per_second": 666.817, | |
| "eval_steps_per_second": 27.784, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 1.2529166666666667e-05, | |
| "loss": 0.335, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.8091715976331361, | |
| "eval_f1": 0.8160105785088146, | |
| "eval_loss": 0.6432847380638123, | |
| "eval_precision": 0.8271351948602929, | |
| "eval_recall": 0.8091715976331361, | |
| "eval_runtime": 24.2377, | |
| "eval_samples_per_second": 669.372, | |
| "eval_steps_per_second": 27.89, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 1.0445833333333334e-05, | |
| "loss": 0.2724, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.8155818540433925, | |
| "eval_f1": 0.8200452160739635, | |
| "eval_loss": 0.778809666633606, | |
| "eval_precision": 0.8263415739712193, | |
| "eval_recall": 0.8155818540433925, | |
| "eval_runtime": 24.3292, | |
| "eval_samples_per_second": 666.853, | |
| "eval_steps_per_second": 27.786, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 8.362500000000002e-06, | |
| "loss": 0.2404, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "eval_accuracy": 0.8102194280078896, | |
| "eval_f1": 0.8165639264989629, | |
| "eval_loss": 0.7925969958305359, | |
| "eval_precision": 0.8272230495650681, | |
| "eval_recall": 0.8102194280078896, | |
| "eval_runtime": 24.3267, | |
| "eval_samples_per_second": 666.921, | |
| "eval_steps_per_second": 27.788, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 6.2791666666666665e-06, | |
| "loss": 0.2417, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_accuracy": 0.811698717948718, | |
| "eval_f1": 0.8176421251758528, | |
| "eval_loss": 0.8324716687202454, | |
| "eval_precision": 0.8284127391855651, | |
| "eval_recall": 0.811698717948718, | |
| "eval_runtime": 24.2309, | |
| "eval_samples_per_second": 669.559, | |
| "eval_steps_per_second": 27.898, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 0.2426, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_accuracy": 0.8218071992110454, | |
| "eval_f1": 0.8237488575881482, | |
| "eval_loss": 0.835399866104126, | |
| "eval_precision": 0.8261715829036445, | |
| "eval_recall": 0.8218071992110454, | |
| "eval_runtime": 24.2418, | |
| "eval_samples_per_second": 669.257, | |
| "eval_steps_per_second": 27.886, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 2.1166666666666666e-06, | |
| "loss": 0.2399, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_accuracy": 0.8130547337278107, | |
| "eval_f1": 0.8181997138713092, | |
| "eval_loss": 0.8061773777008057, | |
| "eval_precision": 0.8260171371000359, | |
| "eval_recall": 0.8130547337278107, | |
| "eval_runtime": 24.2269, | |
| "eval_samples_per_second": 669.67, | |
| "eval_steps_per_second": 27.903, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 3.3333333333333334e-08, | |
| "loss": 0.2292, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "eval_accuracy": 0.8134245562130178, | |
| "eval_f1": 0.8183934985407206, | |
| "eval_loss": 0.8132809996604919, | |
| "eval_precision": 0.8258917495347268, | |
| "eval_recall": 0.8134245562130178, | |
| "eval_runtime": 24.2378, | |
| "eval_samples_per_second": 669.369, | |
| "eval_steps_per_second": 27.89, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "step": 12000, | |
| "total_flos": 1.481337892587438e+17, | |
| "train_loss": 0.42010608355204265, | |
| "train_runtime": 3460.7395, | |
| "train_samples_per_second": 166.438, | |
| "train_steps_per_second": 3.467 | |
| } | |
| ], | |
| "max_steps": 12000, | |
| "num_train_epochs": 4, | |
| "total_flos": 1.481337892587438e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |