| { | |
| "best_metric": 0.8199536621312183, | |
| "best_model_checkpoint": "/home/bram/shares/predict/trained/dutch/hebban-reviews/bert-base-dutch-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": 4.792083333333334e-05, | |
| "loss": 0.7371, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.7546844181459567, | |
| "eval_f1": 0.7629915603619231, | |
| "eval_loss": 0.6527217626571655, | |
| "eval_precision": 0.7780177612862748, | |
| "eval_recall": 0.7546844181459567, | |
| "eval_runtime": 24.4866, | |
| "eval_samples_per_second": 662.566, | |
| "eval_steps_per_second": 27.607, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.58375e-05, | |
| "loss": 0.6457, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.7643614398422091, | |
| "eval_f1": 0.7769010779814548, | |
| "eval_loss": 0.6216204166412354, | |
| "eval_precision": 0.8017754901402682, | |
| "eval_recall": 0.7643614398422091, | |
| "eval_runtime": 24.6912, | |
| "eval_samples_per_second": 657.077, | |
| "eval_steps_per_second": 27.378, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.375416666666667e-05, | |
| "loss": 0.6184, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.7790310650887574, | |
| "eval_f1": 0.789661025567087, | |
| "eval_loss": 0.6005069017410278, | |
| "eval_precision": 0.8094066060995802, | |
| "eval_recall": 0.7790310650887574, | |
| "eval_runtime": 24.5988, | |
| "eval_samples_per_second": 659.544, | |
| "eval_steps_per_second": 27.481, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.1670833333333334e-05, | |
| "loss": 0.6042, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_accuracy": 0.8077539447731755, | |
| "eval_f1": 0.8118714417209679, | |
| "eval_loss": 0.5989466309547424, | |
| "eval_precision": 0.8220143498287384, | |
| "eval_recall": 0.8077539447731755, | |
| "eval_runtime": 24.5976, | |
| "eval_samples_per_second": 659.577, | |
| "eval_steps_per_second": 27.482, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.959166666666667e-05, | |
| "loss": 0.5744, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.7972140039447732, | |
| "eval_f1": 0.8043419281165946, | |
| "eval_loss": 0.5735336542129517, | |
| "eval_precision": 0.8156333669948874, | |
| "eval_recall": 0.7972140039447732, | |
| "eval_runtime": 24.5948, | |
| "eval_samples_per_second": 659.651, | |
| "eval_steps_per_second": 27.485, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.7512500000000004e-05, | |
| "loss": 0.5833, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.8060281065088757, | |
| "eval_f1": 0.8074379985793598, | |
| "eval_loss": 0.5610195994377136, | |
| "eval_precision": 0.8123421315309852, | |
| "eval_recall": 0.8060281065088757, | |
| "eval_runtime": 24.5935, | |
| "eval_samples_per_second": 659.686, | |
| "eval_steps_per_second": 27.487, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.542916666666667e-05, | |
| "loss": 0.4633, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.8055966469428008, | |
| "eval_f1": 0.8112577676458685, | |
| "eval_loss": 0.5967000126838684, | |
| "eval_precision": 0.8196531984723394, | |
| "eval_recall": 0.8055966469428008, | |
| "eval_runtime": 24.7636, | |
| "eval_samples_per_second": 655.155, | |
| "eval_steps_per_second": 27.298, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.3345833333333335e-05, | |
| "loss": 0.4473, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_accuracy": 0.8108357988165681, | |
| "eval_f1": 0.8144157036046853, | |
| "eval_loss": 0.6390397548675537, | |
| "eval_precision": 0.8190905659244092, | |
| "eval_recall": 0.8108357988165681, | |
| "eval_runtime": 24.7631, | |
| "eval_samples_per_second": 655.169, | |
| "eval_steps_per_second": 27.299, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 3.12625e-05, | |
| "loss": 0.4493, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_accuracy": 0.7642998027613412, | |
| "eval_f1": 0.780188565362406, | |
| "eval_loss": 0.6074336171150208, | |
| "eval_precision": 0.818173356753982, | |
| "eval_recall": 0.7642998027613412, | |
| "eval_runtime": 24.6961, | |
| "eval_samples_per_second": 656.945, | |
| "eval_steps_per_second": 27.373, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.917916666666667e-05, | |
| "loss": 0.4402, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_accuracy": 0.817430966469428, | |
| "eval_f1": 0.818650395547064, | |
| "eval_loss": 0.6175166368484497, | |
| "eval_precision": 0.8200081119899918, | |
| "eval_recall": 0.817430966469428, | |
| "eval_runtime": 24.5978, | |
| "eval_samples_per_second": 659.57, | |
| "eval_steps_per_second": 27.482, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 2.7100000000000005e-05, | |
| "loss": 0.4406, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_accuracy": 0.8167529585798816, | |
| "eval_f1": 0.8194587497642177, | |
| "eval_loss": 0.6328938603401184, | |
| "eval_precision": 0.8229261899048395, | |
| "eval_recall": 0.8167529585798816, | |
| "eval_runtime": 24.611, | |
| "eval_samples_per_second": 659.218, | |
| "eval_steps_per_second": 27.467, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.5016666666666667e-05, | |
| "loss": 0.4457, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.8178007889546351, | |
| "eval_f1": 0.8196437981366307, | |
| "eval_loss": 0.6221566796302795, | |
| "eval_precision": 0.8219923397562409, | |
| "eval_recall": 0.8178007889546351, | |
| "eval_runtime": 24.6955, | |
| "eval_samples_per_second": 656.961, | |
| "eval_steps_per_second": 27.373, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.2933333333333333e-05, | |
| "loss": 0.3013, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.8099112426035503, | |
| "eval_f1": 0.8154916160650456, | |
| "eval_loss": 0.7676593661308289, | |
| "eval_precision": 0.8240038245839082, | |
| "eval_recall": 0.8099112426035503, | |
| "eval_runtime": 24.7339, | |
| "eval_samples_per_second": 655.941, | |
| "eval_steps_per_second": 27.331, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 2.0854166666666668e-05, | |
| "loss": 0.2875, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_accuracy": 0.8025764299802761, | |
| "eval_f1": 0.8097813881432507, | |
| "eval_loss": 0.7549890279769897, | |
| "eval_precision": 0.8215043597038734, | |
| "eval_recall": 0.8025764299802761, | |
| "eval_runtime": 24.7599, | |
| "eval_samples_per_second": 655.252, | |
| "eval_steps_per_second": 27.302, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.8770833333333333e-05, | |
| "loss": 0.2804, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_accuracy": 0.8102194280078896, | |
| "eval_f1": 0.8148870229811181, | |
| "eval_loss": 0.804236114025116, | |
| "eval_precision": 0.8221239914552869, | |
| "eval_recall": 0.8102194280078896, | |
| "eval_runtime": 24.6532, | |
| "eval_samples_per_second": 658.09, | |
| "eval_steps_per_second": 27.42, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 1.66875e-05, | |
| "loss": 0.2784, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_accuracy": 0.8136711045364892, | |
| "eval_f1": 0.8173932526970648, | |
| "eval_loss": 0.8103494644165039, | |
| "eval_precision": 0.8223803033268108, | |
| "eval_recall": 0.8136711045364892, | |
| "eval_runtime": 24.5862, | |
| "eval_samples_per_second": 659.883, | |
| "eval_steps_per_second": 27.495, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.4604166666666666e-05, | |
| "loss": 0.275, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_accuracy": 0.8132396449704142, | |
| "eval_f1": 0.8172151251977423, | |
| "eval_loss": 0.8132022619247437, | |
| "eval_precision": 0.8227088066747076, | |
| "eval_recall": 0.8132396449704142, | |
| "eval_runtime": 24.5279, | |
| "eval_samples_per_second": 661.452, | |
| "eval_steps_per_second": 27.56, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 1.2520833333333334e-05, | |
| "loss": 0.2651, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.8065212031558185, | |
| "eval_f1": 0.8120431442321697, | |
| "eval_loss": 0.7826108336448669, | |
| "eval_precision": 0.8202960014382004, | |
| "eval_recall": 0.8065212031558185, | |
| "eval_runtime": 24.7019, | |
| "eval_samples_per_second": 656.791, | |
| "eval_steps_per_second": 27.366, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 1.0441666666666667e-05, | |
| "loss": 0.1855, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.8140409270216963, | |
| "eval_f1": 0.8173959290765784, | |
| "eval_loss": 1.0370513200759888, | |
| "eval_precision": 0.8233538681349268, | |
| "eval_recall": 0.8140409270216963, | |
| "eval_runtime": 24.6749, | |
| "eval_samples_per_second": 657.51, | |
| "eval_steps_per_second": 27.396, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 8.358333333333333e-06, | |
| "loss": 0.1571, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "eval_accuracy": 0.8128081854043393, | |
| "eval_f1": 0.8173020798805183, | |
| "eval_loss": 1.036438226699829, | |
| "eval_precision": 0.8246320290318612, | |
| "eval_recall": 0.8128081854043393, | |
| "eval_runtime": 24.6314, | |
| "eval_samples_per_second": 658.672, | |
| "eval_steps_per_second": 27.445, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 6.275e-06, | |
| "loss": 0.1608, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_accuracy": 0.8108357988165681, | |
| "eval_f1": 0.8154023680396078, | |
| "eval_loss": 1.0521161556243896, | |
| "eval_precision": 0.8221458981635473, | |
| "eval_recall": 0.8108357988165681, | |
| "eval_runtime": 24.6401, | |
| "eval_samples_per_second": 658.44, | |
| "eval_steps_per_second": 27.435, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 4.191666666666667e-06, | |
| "loss": 0.1542, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_accuracy": 0.8197115384615384, | |
| "eval_f1": 0.8199536621312183, | |
| "eval_loss": 1.108322262763977, | |
| "eval_precision": 0.8205846609719522, | |
| "eval_recall": 0.8197115384615384, | |
| "eval_runtime": 24.5935, | |
| "eval_samples_per_second": 659.687, | |
| "eval_steps_per_second": 27.487, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 2.108333333333333e-06, | |
| "loss": 0.1561, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_accuracy": 0.8138560157790927, | |
| "eval_f1": 0.8166825272884544, | |
| "eval_loss": 1.0635404586791992, | |
| "eval_precision": 0.8205938348020483, | |
| "eval_recall": 0.8138560157790927, | |
| "eval_runtime": 24.8393, | |
| "eval_samples_per_second": 653.159, | |
| "eval_steps_per_second": 27.215, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 2.916666666666667e-08, | |
| "loss": 0.1415, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "eval_accuracy": 0.815396942800789, | |
| "eval_f1": 0.8176978993348879, | |
| "eval_loss": 1.0898783206939697, | |
| "eval_precision": 0.8208248888308868, | |
| "eval_recall": 0.815396942800789, | |
| "eval_runtime": 24.6997, | |
| "eval_samples_per_second": 656.85, | |
| "eval_steps_per_second": 27.369, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "step": 12000, | |
| "total_flos": 1.4832731619013427e+17, | |
| "train_loss": 0.37884856541951495, | |
| "train_runtime": 3446.9333, | |
| "train_samples_per_second": 167.105, | |
| "train_steps_per_second": 3.481 | |
| } | |
| ], | |
| "max_steps": 12000, | |
| "num_train_epochs": 4, | |
| "total_flos": 1.4832731619013427e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |