| { | |
| "best_metric": 0.6894944310188293, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_wnli/checkpoint-5", | |
| "epoch": 6.0, | |
| "global_step": 30, | |
| "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": 0.6956, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6894944310188293, | |
| "eval_runtime": 0.1314, | |
| "eval_samples_per_second": 540.311, | |
| "eval_steps_per_second": 7.61, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6944, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.6944735050201416, | |
| "eval_runtime": 0.1308, | |
| "eval_samples_per_second": 542.667, | |
| "eval_steps_per_second": 7.643, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6937, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.6946936845779419, | |
| "eval_runtime": 0.1297, | |
| "eval_samples_per_second": 547.408, | |
| "eval_steps_per_second": 7.71, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.693, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6914201378822327, | |
| "eval_runtime": 0.1315, | |
| "eval_samples_per_second": 540.072, | |
| "eval_steps_per_second": 7.607, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.693, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6898038387298584, | |
| "eval_runtime": 0.1293, | |
| "eval_samples_per_second": 549.119, | |
| "eval_steps_per_second": 7.734, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6932, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6900650858879089, | |
| "eval_runtime": 0.1293, | |
| "eval_samples_per_second": 549.18, | |
| "eval_steps_per_second": 7.735, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 30, | |
| "total_flos": 119459738025984.0, | |
| "train_loss": 0.6938152154286702, | |
| "train_runtime": 47.698, | |
| "train_samples_per_second": 665.646, | |
| "train_steps_per_second": 5.241 | |
| } | |
| ], | |
| "max_steps": 250, | |
| "num_train_epochs": 50, | |
| "total_flos": 119459738025984.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |