| { | |
| "best_metric": 0.9615830183029175, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_qnli_256/checkpoint-1638", | |
| "epoch": 7.0, | |
| "global_step": 5733, | |
| "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": 1.1009, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5987552626761853, | |
| "eval_loss": 0.9864959716796875, | |
| "eval_runtime": 9.8348, | |
| "eval_samples_per_second": 555.476, | |
| "eval_steps_per_second": 4.372, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.019, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6163280248947465, | |
| "eval_loss": 0.9615830183029175, | |
| "eval_runtime": 9.7787, | |
| "eval_samples_per_second": 558.662, | |
| "eval_steps_per_second": 4.397, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.9743, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.613399231191653, | |
| "eval_loss": 0.9671955704689026, | |
| "eval_runtime": 9.8271, | |
| "eval_samples_per_second": 555.911, | |
| "eval_steps_per_second": 4.376, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.942, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6069924949661358, | |
| "eval_loss": 0.9723971486091614, | |
| "eval_runtime": 9.779, | |
| "eval_samples_per_second": 558.647, | |
| "eval_steps_per_second": 4.397, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.9189, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6016840563792788, | |
| "eval_loss": 0.9827193021774292, | |
| "eval_runtime": 9.8296, | |
| "eval_samples_per_second": 555.769, | |
| "eval_steps_per_second": 4.375, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.898, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5958264689730917, | |
| "eval_loss": 1.0090252161026, | |
| "eval_runtime": 9.7883, | |
| "eval_samples_per_second": 558.113, | |
| "eval_steps_per_second": 4.393, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.8798, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5967417170053084, | |
| "eval_loss": 1.0317091941833496, | |
| "eval_runtime": 9.8635, | |
| "eval_samples_per_second": 553.859, | |
| "eval_steps_per_second": 4.359, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 5733, | |
| "total_flos": 1.931335714419507e+16, | |
| "train_loss": 0.9618304146328003, | |
| "train_runtime": 3166.4896, | |
| "train_samples_per_second": 1653.929, | |
| "train_steps_per_second": 12.932 | |
| } | |
| ], | |
| "max_steps": 40950, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.931335714419507e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |