| { | |
| "best_metric": 0.41574952006340027, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_sst2/checkpoint-527", | |
| "epoch": 6.0, | |
| "global_step": 3162, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9003795066413664e-05, | |
| "loss": 0.487, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8027522935779816, | |
| "eval_loss": 0.41574952006340027, | |
| "eval_runtime": 1.4703, | |
| "eval_samples_per_second": 593.065, | |
| "eval_steps_per_second": 4.761, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.800379506641367e-05, | |
| "loss": 0.2824, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8004587155963303, | |
| "eval_loss": 0.43506258726119995, | |
| "eval_runtime": 1.416, | |
| "eval_samples_per_second": 615.81, | |
| "eval_steps_per_second": 4.943, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.700379506641366e-05, | |
| "loss": 0.2265, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8096330275229358, | |
| "eval_loss": 0.4486955404281616, | |
| "eval_runtime": 1.4216, | |
| "eval_samples_per_second": 613.408, | |
| "eval_steps_per_second": 4.924, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.6003795066413666e-05, | |
| "loss": 0.1989, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7993119266055045, | |
| "eval_loss": 0.5181797742843628, | |
| "eval_runtime": 1.4169, | |
| "eval_samples_per_second": 615.418, | |
| "eval_steps_per_second": 4.94, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.50056925996205e-05, | |
| "loss": 0.1813, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7981651376146789, | |
| "eval_loss": 0.46536529064178467, | |
| "eval_runtime": 1.4169, | |
| "eval_samples_per_second": 615.449, | |
| "eval_steps_per_second": 4.941, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4005692599620494e-05, | |
| "loss": 0.1684, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7924311926605505, | |
| "eval_loss": 0.5339637398719788, | |
| "eval_runtime": 1.4148, | |
| "eval_samples_per_second": 616.334, | |
| "eval_steps_per_second": 4.948, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 3162, | |
| "total_flos": 1.2670068697595904e+16, | |
| "train_loss": 0.2574223995510345, | |
| "train_runtime": 2226.6844, | |
| "train_samples_per_second": 1512.316, | |
| "train_steps_per_second": 11.834 | |
| } | |
| ], | |
| "max_steps": 26350, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.2670068697595904e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |