| { | |
| "best_metric": 1.952246904373169, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_data_aug_wnli_128/checkpoint-435", | |
| "epoch": 6.0, | |
| "global_step": 2610, | |
| "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": 0.6679, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.14084507042253522, | |
| "eval_loss": 1.952246904373169, | |
| "eval_runtime": 0.1404, | |
| "eval_samples_per_second": 505.649, | |
| "eval_steps_per_second": 7.122, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.4943, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.11267605633802817, | |
| "eval_loss": 3.652181386947632, | |
| "eval_runtime": 0.1331, | |
| "eval_samples_per_second": 533.553, | |
| "eval_steps_per_second": 7.515, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.4405, | |
| "step": 1305 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.1267605633802817, | |
| "eval_loss": 4.511776924133301, | |
| "eval_runtime": 0.1324, | |
| "eval_samples_per_second": 536.125, | |
| "eval_steps_per_second": 7.551, | |
| "step": 1305 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.4086, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.11267605633802817, | |
| "eval_loss": 6.04368782043457, | |
| "eval_runtime": 0.1327, | |
| "eval_samples_per_second": 535.177, | |
| "eval_steps_per_second": 7.538, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.3818, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.11267605633802817, | |
| "eval_loss": 9.593058586120605, | |
| "eval_runtime": 0.1318, | |
| "eval_samples_per_second": 538.611, | |
| "eval_steps_per_second": 7.586, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.3492, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.14084507042253522, | |
| "eval_loss": 6.966458797454834, | |
| "eval_runtime": 0.1319, | |
| "eval_samples_per_second": 538.276, | |
| "eval_steps_per_second": 7.581, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 2610, | |
| "total_flos": 7956940701302784.0, | |
| "train_loss": 0.45706886496123683, | |
| "train_runtime": 1646.7328, | |
| "train_samples_per_second": 1689.406, | |
| "train_steps_per_second": 13.208 | |
| } | |
| ], | |
| "max_steps": 21750, | |
| "num_train_epochs": 50, | |
| "total_flos": 7956940701302784.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |