| { | |
| "best_metric": 0.8079156875610352, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_qqp/checkpoint-17058", | |
| "epoch": 11.0, | |
| "global_step": 31273, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 1.2837, | |
| "step": 2843 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 1.2201430797576904, | |
| "eval_runtime": 89.092, | |
| "eval_samples_per_second": 453.801, | |
| "eval_steps_per_second": 3.547, | |
| "step": 2843 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.076, | |
| "step": 5686 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7442740539203562, | |
| "eval_combined_score": 0.66490110793833, | |
| "eval_f1": 0.585528161956304, | |
| "eval_loss": 0.847718358039856, | |
| "eval_runtime": 89.222, | |
| "eval_samples_per_second": 453.139, | |
| "eval_steps_per_second": 3.542, | |
| "step": 5686 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.866, | |
| "step": 8529 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7518179569626515, | |
| "eval_combined_score": 0.6721315617469634, | |
| "eval_f1": 0.5924451665312754, | |
| "eval_loss": 0.8216822743415833, | |
| "eval_runtime": 89.3998, | |
| "eval_samples_per_second": 452.238, | |
| "eval_steps_per_second": 3.535, | |
| "step": 8529 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.8317, | |
| "step": 11372 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7564927034380411, | |
| "eval_combined_score": 0.6903861830360454, | |
| "eval_f1": 0.6242796626340495, | |
| "eval_loss": 0.8135799765586853, | |
| "eval_runtime": 89.2169, | |
| "eval_samples_per_second": 453.165, | |
| "eval_steps_per_second": 3.542, | |
| "step": 11372 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.8122, | |
| "step": 14215 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7587682414048974, | |
| "eval_combined_score": 0.696969127996804, | |
| "eval_f1": 0.6351700145887106, | |
| "eval_loss": 0.8125847578048706, | |
| "eval_runtime": 89.3981, | |
| "eval_samples_per_second": 452.247, | |
| "eval_steps_per_second": 3.535, | |
| "step": 14215 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.799, | |
| "step": 17058 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.756987385604749, | |
| "eval_combined_score": 0.6809586089630565, | |
| "eval_f1": 0.604929832321364, | |
| "eval_loss": 0.8079156875610352, | |
| "eval_runtime": 89.2539, | |
| "eval_samples_per_second": 452.977, | |
| "eval_steps_per_second": 3.54, | |
| "step": 17058 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 3.8658113487167835e+20, | |
| "step": 19901 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 89.0992, | |
| "eval_samples_per_second": 453.764, | |
| "eval_steps_per_second": 3.547, | |
| "step": 19901 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.0, | |
| "step": 22744 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 89.2111, | |
| "eval_samples_per_second": 453.195, | |
| "eval_steps_per_second": 3.542, | |
| "step": 22744 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.0, | |
| "step": 25587 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 89.3622, | |
| "eval_samples_per_second": 452.428, | |
| "eval_steps_per_second": 3.536, | |
| "step": 25587 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 28430 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 89.0175, | |
| "eval_samples_per_second": 454.18, | |
| "eval_steps_per_second": 3.55, | |
| "step": 28430 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.0, | |
| "step": 31273 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 89.0784, | |
| "eval_samples_per_second": 453.87, | |
| "eval_steps_per_second": 3.547, | |
| "step": 31273 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "step": 31273, | |
| "total_flos": 1.1838967078138675e+17, | |
| "train_loss": 3.514373953378894e+19, | |
| "train_runtime": 23780.3759, | |
| "train_samples_per_second": 765.013, | |
| "train_steps_per_second": 5.978 | |
| } | |
| ], | |
| "max_steps": 142150, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.1838967078138675e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |