| { | |
| "best_metric": 0.9071565051241967, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/mbert/mbert-base-finetuned-pos-ud-Vietnamese-VTB/checkpoint-1000", | |
| "epoch": 79.54545454545455, | |
| "global_step": 3500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 7.92e-05, | |
| "loss": 0.9833, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 7.946845637583894e-05, | |
| "loss": 0.1592, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 7.893154362416109e-05, | |
| "loss": 0.0706, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 7.839463087248322e-05, | |
| "loss": 0.0405, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "learning_rate": 7.785771812080537e-05, | |
| "loss": 0.0298, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "eval_accuracy": 0.9025534132360604, | |
| "eval_loss": 0.5368738174438477, | |
| "eval_runtime": 4.0641, | |
| "eval_samples_per_second": 196.846, | |
| "eval_steps_per_second": 24.606, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "learning_rate": 7.73261744966443e-05, | |
| "loss": 0.0239, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 7.678926174496645e-05, | |
| "loss": 0.0189, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 7.62523489932886e-05, | |
| "loss": 0.0159, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 20.45, | |
| "learning_rate": 7.571543624161075e-05, | |
| "loss": 0.0143, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "learning_rate": 7.51785234899329e-05, | |
| "loss": 0.012, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "eval_accuracy": 0.9071565051241967, | |
| "eval_loss": 0.6289473176002502, | |
| "eval_runtime": 4.0794, | |
| "eval_samples_per_second": 196.105, | |
| "eval_steps_per_second": 24.513, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 7.464161073825505e-05, | |
| "loss": 0.0116, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 7.410469798657718e-05, | |
| "loss": 0.0088, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 29.55, | |
| "learning_rate": 7.356778523489933e-05, | |
| "loss": 0.011, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 31.82, | |
| "learning_rate": 7.303087248322148e-05, | |
| "loss": 0.0102, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 34.09, | |
| "learning_rate": 7.249395973154363e-05, | |
| "loss": 0.0094, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 34.09, | |
| "eval_accuracy": 0.9035956227201668, | |
| "eval_loss": 0.6397399306297302, | |
| "eval_runtime": 4.065, | |
| "eval_samples_per_second": 196.802, | |
| "eval_steps_per_second": 24.6, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "learning_rate": 7.195704697986577e-05, | |
| "loss": 0.0079, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 38.64, | |
| "learning_rate": 7.142013422818792e-05, | |
| "loss": 0.0083, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 40.91, | |
| "learning_rate": 7.088322147651007e-05, | |
| "loss": 0.008, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 43.18, | |
| "learning_rate": 7.034630872483222e-05, | |
| "loss": 0.0079, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "learning_rate": 6.980939597315437e-05, | |
| "loss": 0.007, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "eval_accuracy": 0.9045509814139309, | |
| "eval_loss": 0.7602550983428955, | |
| "eval_runtime": 4.0644, | |
| "eval_samples_per_second": 196.83, | |
| "eval_steps_per_second": 24.604, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 47.73, | |
| "learning_rate": 6.927248322147651e-05, | |
| "loss": 0.0057, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 6.873557046979866e-05, | |
| "loss": 0.0064, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 52.27, | |
| "learning_rate": 6.819865771812081e-05, | |
| "loss": 0.0059, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 54.55, | |
| "learning_rate": 6.766174496644296e-05, | |
| "loss": 0.0046, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 56.82, | |
| "learning_rate": 6.712483221476511e-05, | |
| "loss": 0.0066, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 56.82, | |
| "eval_accuracy": 0.8987319784610039, | |
| "eval_loss": 0.7796401381492615, | |
| "eval_runtime": 4.0558, | |
| "eval_samples_per_second": 197.249, | |
| "eval_steps_per_second": 24.656, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 59.09, | |
| "learning_rate": 6.658791946308726e-05, | |
| "loss": 0.0059, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 61.36, | |
| "learning_rate": 6.60510067114094e-05, | |
| "loss": 0.0051, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 63.64, | |
| "learning_rate": 6.551409395973155e-05, | |
| "loss": 0.0072, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 65.91, | |
| "learning_rate": 6.49771812080537e-05, | |
| "loss": 0.0042, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "learning_rate": 6.444026845637585e-05, | |
| "loss": 0.0045, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "eval_accuracy": 0.9019454577036651, | |
| "eval_loss": 0.7631990313529968, | |
| "eval_runtime": 4.0599, | |
| "eval_samples_per_second": 197.049, | |
| "eval_steps_per_second": 24.631, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 70.45, | |
| "learning_rate": 6.3903355704698e-05, | |
| "loss": 0.0045, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 72.73, | |
| "learning_rate": 6.337181208053692e-05, | |
| "loss": 0.0055, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 6.283489932885907e-05, | |
| "loss": 0.0054, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 77.27, | |
| "learning_rate": 6.229798657718121e-05, | |
| "loss": 0.0038, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 79.55, | |
| "learning_rate": 6.176107382550336e-05, | |
| "loss": 0.0056, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 79.55, | |
| "eval_accuracy": 0.9014243529616119, | |
| "eval_loss": 0.7848494648933411, | |
| "eval_runtime": 4.0549, | |
| "eval_samples_per_second": 197.294, | |
| "eval_steps_per_second": 24.662, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 79.55, | |
| "step": 3500, | |
| "total_flos": 1.4552021793583104e+16, | |
| "train_loss": 0.04398087625844138, | |
| "train_runtime": 970.2394, | |
| "train_samples_per_second": 494.723, | |
| "train_steps_per_second": 15.46 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 341, | |
| "total_flos": 1.4552021793583104e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |