| { | |
| "best_metric": 82.48871048670347, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/mbert/bert-base-finetuned-parsing-ud-Korean-GSD/checkpoint-2000", | |
| "epoch": 32.608695652173914, | |
| "global_step": 4500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 7.76e-05, | |
| "loss": 3.3939, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 7.947919463087248e-05, | |
| "loss": 1.0824, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 7.894228187919463e-05, | |
| "loss": 0.8005, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 7.840536912751678e-05, | |
| "loss": 0.5864, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 7.786845637583893e-05, | |
| "loss": 0.4237, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_las": 80.40642247867537, | |
| "eval_loss": 0.9098409414291382, | |
| "eval_runtime": 6.054, | |
| "eval_samples_per_second": 156.921, | |
| "eval_steps_per_second": 19.656, | |
| "eval_uas": 84.83860177287171, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 7.733154362416108e-05, | |
| "loss": 0.3544, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 7.679463087248322e-05, | |
| "loss": 0.2966, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 7.625771812080537e-05, | |
| "loss": 0.2265, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 7.572080536912752e-05, | |
| "loss": 0.1954, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 7.518389261744967e-05, | |
| "loss": 0.1759, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "eval_las": 81.9284161230975, | |
| "eval_loss": 1.3141283988952637, | |
| "eval_runtime": 6.0488, | |
| "eval_samples_per_second": 157.057, | |
| "eval_steps_per_second": 19.673, | |
| "eval_uas": 86.06790433182806, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 7.464697986577182e-05, | |
| "loss": 0.1628, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 7.411006711409397e-05, | |
| "loss": 0.1266, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 7.357315436241611e-05, | |
| "loss": 0.1255, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 7.303624161073826e-05, | |
| "loss": 0.1183, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 7.249932885906041e-05, | |
| "loss": 0.1053, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "eval_las": 81.93677872553938, | |
| "eval_loss": 1.3510112762451172, | |
| "eval_runtime": 6.0457, | |
| "eval_samples_per_second": 157.137, | |
| "eval_steps_per_second": 19.684, | |
| "eval_uas": 86.15989295868874, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 7.196241610738256e-05, | |
| "loss": 0.0971, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 7.142550335570471e-05, | |
| "loss": 0.0906, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 7.088859060402686e-05, | |
| "loss": 0.0805, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 7.0351677852349e-05, | |
| "loss": 0.077, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 6.981476510067114e-05, | |
| "loss": 0.0688, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "eval_las": 82.48871048670347, | |
| "eval_loss": 1.5172237157821655, | |
| "eval_runtime": 6.0437, | |
| "eval_samples_per_second": 157.189, | |
| "eval_steps_per_second": 19.69, | |
| "eval_uas": 86.62819869543402, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "learning_rate": 6.927785234899329e-05, | |
| "loss": 0.0692, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 15.94, | |
| "learning_rate": 6.874093959731543e-05, | |
| "loss": 0.0678, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 6.820402684563758e-05, | |
| "loss": 0.0587, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "learning_rate": 6.766711409395973e-05, | |
| "loss": 0.0621, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 6.713020134228188e-05, | |
| "loss": 0.0579, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "eval_las": 82.36327145007526, | |
| "eval_loss": 1.7050881385803223, | |
| "eval_runtime": 6.0527, | |
| "eval_samples_per_second": 156.956, | |
| "eval_steps_per_second": 19.661, | |
| "eval_uas": 86.70346211741094, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 6.659328859060403e-05, | |
| "loss": 0.0518, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 19.57, | |
| "learning_rate": 6.605637583892618e-05, | |
| "loss": 0.0529, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 20.29, | |
| "learning_rate": 6.551946308724832e-05, | |
| "loss": 0.0519, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 6.498255033557047e-05, | |
| "loss": 0.0491, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 6.444563758389262e-05, | |
| "loss": 0.0438, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "eval_las": 82.26292022077271, | |
| "eval_loss": 1.7979624271392822, | |
| "eval_runtime": 6.0527, | |
| "eval_samples_per_second": 156.955, | |
| "eval_steps_per_second": 19.661, | |
| "eval_uas": 86.36895801973574, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 22.46, | |
| "learning_rate": 6.390872483221477e-05, | |
| "loss": 0.045, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 23.19, | |
| "learning_rate": 6.337181208053692e-05, | |
| "loss": 0.0457, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 23.91, | |
| "learning_rate": 6.284026845637584e-05, | |
| "loss": 0.0417, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 24.64, | |
| "learning_rate": 6.230335570469799e-05, | |
| "loss": 0.0381, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 25.36, | |
| "learning_rate": 6.176644295302013e-05, | |
| "loss": 0.0387, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 25.36, | |
| "eval_las": 81.73607626693426, | |
| "eval_loss": 1.8726658821105957, | |
| "eval_runtime": 6.0404, | |
| "eval_samples_per_second": 157.274, | |
| "eval_steps_per_second": 19.701, | |
| "eval_uas": 85.83375146345543, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 26.09, | |
| "learning_rate": 6.122953020134228e-05, | |
| "loss": 0.038, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 26.81, | |
| "learning_rate": 6.069261744966444e-05, | |
| "loss": 0.0367, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 27.54, | |
| "learning_rate": 6.0155704697986585e-05, | |
| "loss": 0.0361, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 28.26, | |
| "learning_rate": 5.9618791946308734e-05, | |
| "loss": 0.0396, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "learning_rate": 5.9081879194630875e-05, | |
| "loss": 0.0373, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "eval_las": 82.27964542565647, | |
| "eval_loss": 1.884783387184143, | |
| "eval_runtime": 6.0511, | |
| "eval_samples_per_second": 156.995, | |
| "eval_steps_per_second": 19.666, | |
| "eval_uas": 86.27696939287506, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 29.71, | |
| "learning_rate": 5.854496644295302e-05, | |
| "loss": 0.0296, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 30.43, | |
| "learning_rate": 5.800805369127517e-05, | |
| "loss": 0.0291, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 31.16, | |
| "learning_rate": 5.747114093959732e-05, | |
| "loss": 0.0301, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 31.88, | |
| "learning_rate": 5.693422818791947e-05, | |
| "loss": 0.0321, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 32.61, | |
| "learning_rate": 5.6397315436241616e-05, | |
| "loss": 0.0281, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 32.61, | |
| "eval_las": 82.12075597926075, | |
| "eval_loss": 2.1597061157226562, | |
| "eval_runtime": 6.0549, | |
| "eval_samples_per_second": 156.896, | |
| "eval_steps_per_second": 19.653, | |
| "eval_uas": 85.9675531025255, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 32.61, | |
| "step": 4500, | |
| "total_flos": 2.3959177620750336e+16, | |
| "train_loss": 0.21554310941696167, | |
| "train_runtime": 2399.8621, | |
| "train_samples_per_second": 200.011, | |
| "train_steps_per_second": 6.25 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 109, | |
| "total_flos": 2.3959177620750336e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |