| { | |
| "best_metric": 0.9700717616323291, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/mbert/mbert-base-finetuned-pos-ud-Arabic-PADT/checkpoint-2500", | |
| "epoch": 26.31578947368421, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 7.840000000000001e-05, | |
| "loss": 0.9117, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 7.947382550335571e-05, | |
| "loss": 0.154, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 7.893691275167786e-05, | |
| "loss": 0.1201, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 7.840000000000001e-05, | |
| "loss": 0.1031, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 7.786308724832216e-05, | |
| "loss": 0.0781, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_accuracy": 0.9692119448394457, | |
| "eval_loss": 0.11606918275356293, | |
| "eval_runtime": 4.6573, | |
| "eval_samples_per_second": 195.177, | |
| "eval_steps_per_second": 24.478, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 7.73261744966443e-05, | |
| "loss": 0.0712, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 7.678926174496645e-05, | |
| "loss": 0.0585, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 7.62523489932886e-05, | |
| "loss": 0.053, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 7.571543624161075e-05, | |
| "loss": 0.0439, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 7.51785234899329e-05, | |
| "loss": 0.0393, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "eval_accuracy": 0.9695426436059393, | |
| "eval_loss": 0.12902098894119263, | |
| "eval_runtime": 4.6945, | |
| "eval_samples_per_second": 193.631, | |
| "eval_steps_per_second": 24.284, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 7.464161073825505e-05, | |
| "loss": 0.0338, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 7.410469798657718e-05, | |
| "loss": 0.03, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 7.356778523489933e-05, | |
| "loss": 0.0272, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 7.303087248322148e-05, | |
| "loss": 0.023, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 7.249395973154363e-05, | |
| "loss": 0.0233, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "eval_accuracy": 0.9680544991567182, | |
| "eval_loss": 0.1491442620754242, | |
| "eval_runtime": 4.6841, | |
| "eval_samples_per_second": 194.061, | |
| "eval_steps_per_second": 24.338, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 7.195704697986577e-05, | |
| "loss": 0.0196, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 7.142013422818792e-05, | |
| "loss": 0.019, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 7.088322147651007e-05, | |
| "loss": 0.0148, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.034630872483222e-05, | |
| "loss": 0.0165, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "learning_rate": 6.980939597315437e-05, | |
| "loss": 0.0143, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "eval_accuracy": 0.9679222196501207, | |
| "eval_loss": 0.17728525400161743, | |
| "eval_runtime": 4.6951, | |
| "eval_samples_per_second": 193.605, | |
| "eval_steps_per_second": 24.28, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 6.927248322147651e-05, | |
| "loss": 0.0135, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "learning_rate": 6.873557046979866e-05, | |
| "loss": 0.0115, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 12.11, | |
| "learning_rate": 6.819865771812081e-05, | |
| "loss": 0.0127, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "learning_rate": 6.766174496644296e-05, | |
| "loss": 0.0109, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 6.712483221476511e-05, | |
| "loss": 0.0122, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "eval_accuracy": 0.9700717616323291, | |
| "eval_loss": 0.1844806671142578, | |
| "eval_runtime": 4.7037, | |
| "eval_samples_per_second": 193.251, | |
| "eval_steps_per_second": 24.236, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "learning_rate": 6.658791946308726e-05, | |
| "loss": 0.0098, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 6.60510067114094e-05, | |
| "loss": 0.0104, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "learning_rate": 6.551409395973155e-05, | |
| "loss": 0.0089, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 15.26, | |
| "learning_rate": 6.49771812080537e-05, | |
| "loss": 0.0092, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 6.444026845637585e-05, | |
| "loss": 0.0086, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "eval_accuracy": 0.9694103640993419, | |
| "eval_loss": 0.18017974495887756, | |
| "eval_runtime": 4.6947, | |
| "eval_samples_per_second": 193.623, | |
| "eval_steps_per_second": 24.283, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.32, | |
| "learning_rate": 6.3903355704698e-05, | |
| "loss": 0.0084, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "learning_rate": 6.336644295302015e-05, | |
| "loss": 0.0082, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 17.37, | |
| "learning_rate": 6.28295302013423e-05, | |
| "loss": 0.0068, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "learning_rate": 6.229261744966444e-05, | |
| "loss": 0.0063, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "learning_rate": 6.175570469798658e-05, | |
| "loss": 0.0059, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "eval_accuracy": 0.9698402724957836, | |
| "eval_loss": 0.20168748497962952, | |
| "eval_runtime": 4.6845, | |
| "eval_samples_per_second": 194.044, | |
| "eval_steps_per_second": 24.336, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 18.95, | |
| "learning_rate": 6.121879194630873e-05, | |
| "loss": 0.0067, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 6.068187919463087e-05, | |
| "loss": 0.0068, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 6.014496644295302e-05, | |
| "loss": 0.0074, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 20.53, | |
| "learning_rate": 5.960805369127517e-05, | |
| "loss": 0.006, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "learning_rate": 5.907114093959732e-05, | |
| "loss": 0.0061, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "eval_accuracy": 0.969873342372433, | |
| "eval_loss": 0.1942814141511917, | |
| "eval_runtime": 4.6772, | |
| "eval_samples_per_second": 194.346, | |
| "eval_steps_per_second": 24.373, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.58, | |
| "learning_rate": 5.8534228187919466e-05, | |
| "loss": 0.0064, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "learning_rate": 5.7997315436241614e-05, | |
| "loss": 0.0053, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 22.63, | |
| "learning_rate": 5.746040268456376e-05, | |
| "loss": 0.0048, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 23.16, | |
| "learning_rate": 5.692348993288591e-05, | |
| "loss": 0.0061, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "learning_rate": 5.638657718120806e-05, | |
| "loss": 0.0052, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "eval_accuracy": 0.9685174774298092, | |
| "eval_loss": 0.2084885686635971, | |
| "eval_runtime": 4.6977, | |
| "eval_samples_per_second": 193.498, | |
| "eval_steps_per_second": 24.267, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 24.21, | |
| "learning_rate": 5.58496644295302e-05, | |
| "loss": 0.0049, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 24.74, | |
| "learning_rate": 5.531275167785235e-05, | |
| "loss": 0.0041, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 25.26, | |
| "learning_rate": 5.47758389261745e-05, | |
| "loss": 0.0054, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 25.79, | |
| "learning_rate": 5.4238926174496645e-05, | |
| "loss": 0.0049, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 5.370738255033558e-05, | |
| "loss": 0.0045, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "eval_accuracy": 0.9700056218790304, | |
| "eval_loss": 0.2086580991744995, | |
| "eval_runtime": 4.685, | |
| "eval_samples_per_second": 194.024, | |
| "eval_steps_per_second": 24.333, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "step": 5000, | |
| "total_flos": 2.088958878798336e+16, | |
| "train_loss": 0.0416443316757679, | |
| "train_runtime": 1404.0367, | |
| "train_samples_per_second": 341.871, | |
| "train_steps_per_second": 10.683 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 79, | |
| "total_flos": 2.088958878798336e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |