| { | |
| "best_metric": 0.6053687907676869, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/bert/bert-base-finetuned-pos-ud-Korean-GSD/checkpoint-500", | |
| "epoch": 21.73913043478261, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.9e-05, | |
| "loss": 1.5142, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.967114093959731e-05, | |
| "loss": 1.0614, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 4.933557046979866e-05, | |
| "loss": 1.042, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 4.9e-05, | |
| "loss": 1.0138, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 4.8664429530201344e-05, | |
| "loss": 1.0024, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_accuracy": 0.6053687907676869, | |
| "eval_loss": 1.007150650024414, | |
| "eval_runtime": 4.8522, | |
| "eval_samples_per_second": 195.787, | |
| "eval_steps_per_second": 24.525, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 4.8328859060402684e-05, | |
| "loss": 1.0071, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 4.799328859060403e-05, | |
| "loss": 0.9899, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 4.7657718120805376e-05, | |
| "loss": 0.9818, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 4.7322147651006715e-05, | |
| "loss": 0.9812, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 4.698657718120806e-05, | |
| "loss": 0.9736, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "eval_accuracy": 0.6013547415955846, | |
| "eval_loss": 1.0191282033920288, | |
| "eval_runtime": 4.8652, | |
| "eval_samples_per_second": 195.263, | |
| "eval_steps_per_second": 24.459, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 4.66510067114094e-05, | |
| "loss": 0.9626, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 4.631543624161074e-05, | |
| "loss": 0.9428, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 4.597986577181208e-05, | |
| "loss": 0.9431, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 4.5644295302013425e-05, | |
| "loss": 0.9438, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 4.5308724832214764e-05, | |
| "loss": 0.9281, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "eval_accuracy": 0.6036126442548921, | |
| "eval_loss": 1.0334581136703491, | |
| "eval_runtime": 4.8675, | |
| "eval_samples_per_second": 195.172, | |
| "eval_steps_per_second": 24.448, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 4.497315436241611e-05, | |
| "loss": 0.9, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 4.463758389261745e-05, | |
| "loss": 0.9157, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 4.4302013422818796e-05, | |
| "loss": 0.8991, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 4.3966442953020135e-05, | |
| "loss": 0.8891, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 4.363087248322148e-05, | |
| "loss": 0.8694, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "eval_accuracy": 0.5921558789095166, | |
| "eval_loss": 1.0879769325256348, | |
| "eval_runtime": 4.8603, | |
| "eval_samples_per_second": 195.461, | |
| "eval_steps_per_second": 24.484, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "learning_rate": 4.329530201342282e-05, | |
| "loss": 0.8726, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 15.94, | |
| "learning_rate": 4.2959731543624166e-05, | |
| "loss": 0.8605, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 4.2624161073825505e-05, | |
| "loss": 0.8449, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "learning_rate": 4.2288590604026845e-05, | |
| "loss": 0.8285, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 4.195302013422819e-05, | |
| "loss": 0.8418, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "eval_accuracy": 0.5876400735909015, | |
| "eval_loss": 1.1507753133773804, | |
| "eval_runtime": 4.8677, | |
| "eval_samples_per_second": 195.164, | |
| "eval_steps_per_second": 24.447, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 4.161744966442953e-05, | |
| "loss": 0.8277, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 19.57, | |
| "learning_rate": 4.1281879194630876e-05, | |
| "loss": 0.8069, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 20.29, | |
| "learning_rate": 4.0946308724832215e-05, | |
| "loss": 0.7859, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 4.061073825503356e-05, | |
| "loss": 0.8126, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 4.02751677852349e-05, | |
| "loss": 0.7752, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "eval_accuracy": 0.5837096504432179, | |
| "eval_loss": 1.1938718557357788, | |
| "eval_runtime": 4.8673, | |
| "eval_samples_per_second": 195.178, | |
| "eval_steps_per_second": 24.449, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "step": 3000, | |
| "total_flos": 1.2500041419988992e+16, | |
| "train_loss": 0.9339236297607422, | |
| "train_runtime": 790.8493, | |
| "train_samples_per_second": 606.942, | |
| "train_steps_per_second": 18.967 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 109, | |
| "total_flos": 1.2500041419988992e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |