| { | |
| "best_metric": 0.9572671015219937, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/mbert/mbert-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": 7.840000000000001e-05, | |
| "loss": 0.9879, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 7.947382550335571e-05, | |
| "loss": 0.2181, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 7.893691275167786e-05, | |
| "loss": 0.1495, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 7.840000000000001e-05, | |
| "loss": 0.1132, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 7.786308724832216e-05, | |
| "loss": 0.0799, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_accuracy": 0.9572671015219937, | |
| "eval_loss": 0.18942660093307495, | |
| "eval_runtime": 4.8972, | |
| "eval_samples_per_second": 193.989, | |
| "eval_steps_per_second": 24.3, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 7.73261744966443e-05, | |
| "loss": 0.0661, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 7.678926174496645e-05, | |
| "loss": 0.0532, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 7.62523489932886e-05, | |
| "loss": 0.0401, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 7.571543624161075e-05, | |
| "loss": 0.0368, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 7.51785234899329e-05, | |
| "loss": 0.0331, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "eval_accuracy": 0.9542565646429169, | |
| "eval_loss": 0.23469237983226776, | |
| "eval_runtime": 4.9192, | |
| "eval_samples_per_second": 193.122, | |
| "eval_steps_per_second": 24.191, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 7.464161073825505e-05, | |
| "loss": 0.0279, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 7.410469798657718e-05, | |
| "loss": 0.0276, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 7.356778523489933e-05, | |
| "loss": 0.0239, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 7.303087248322148e-05, | |
| "loss": 0.0211, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 7.249395973154363e-05, | |
| "loss": 0.019, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "eval_accuracy": 0.9547583207894297, | |
| "eval_loss": 0.2637677788734436, | |
| "eval_runtime": 4.9181, | |
| "eval_samples_per_second": 193.163, | |
| "eval_steps_per_second": 24.196, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 7.195704697986577e-05, | |
| "loss": 0.0174, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 7.142013422818792e-05, | |
| "loss": 0.0178, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 7.088322147651007e-05, | |
| "loss": 0.0181, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 7.034630872483222e-05, | |
| "loss": 0.014, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 6.980939597315437e-05, | |
| "loss": 0.0144, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "eval_accuracy": 0.9535875564475664, | |
| "eval_loss": 0.2897445857524872, | |
| "eval_runtime": 4.9125, | |
| "eval_samples_per_second": 193.386, | |
| "eval_steps_per_second": 24.224, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "learning_rate": 6.927248322147651e-05, | |
| "loss": 0.0154, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 15.94, | |
| "learning_rate": 6.873557046979866e-05, | |
| "loss": 0.0133, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 6.819865771812081e-05, | |
| "loss": 0.0142, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "learning_rate": 6.766174496644296e-05, | |
| "loss": 0.0132, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 6.712483221476511e-05, | |
| "loss": 0.013, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "eval_accuracy": 0.9560127111557116, | |
| "eval_loss": 0.2839347720146179, | |
| "eval_runtime": 4.9076, | |
| "eval_samples_per_second": 193.578, | |
| "eval_steps_per_second": 24.248, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 6.658791946308726e-05, | |
| "loss": 0.0114, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 19.57, | |
| "learning_rate": 6.60510067114094e-05, | |
| "loss": 0.0106, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 20.29, | |
| "learning_rate": 6.551409395973155e-05, | |
| "loss": 0.0103, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 6.49771812080537e-05, | |
| "loss": 0.0123, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 6.444026845637585e-05, | |
| "loss": 0.0093, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "eval_accuracy": 0.9540056865696604, | |
| "eval_loss": 0.2998155355453491, | |
| "eval_runtime": 4.8962, | |
| "eval_samples_per_second": 194.026, | |
| "eval_steps_per_second": 24.304, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "step": 3000, | |
| "total_flos": 1.2500041419988992e+16, | |
| "train_loss": 0.0700680200457573, | |
| "train_runtime": 844.3355, | |
| "train_samples_per_second": 568.494, | |
| "train_steps_per_second": 17.765 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 109, | |
| "total_flos": 1.2500041419988992e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |