| { | |
| "best_metric": 1.7938315868377686, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_mnli/checkpoint-15340", | |
| "epoch": 10.0, | |
| "global_step": 30680, | |
| "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": 1.8866, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.7940564155578613, | |
| "eval_runtime": 21.7409, | |
| "eval_samples_per_second": 451.454, | |
| "eval_steps_per_second": 3.542, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.8864, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.7939273118972778, | |
| "eval_runtime": 21.6617, | |
| "eval_samples_per_second": 453.103, | |
| "eval_steps_per_second": 3.555, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.8864, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.7943607568740845, | |
| "eval_runtime": 21.71, | |
| "eval_samples_per_second": 452.096, | |
| "eval_steps_per_second": 3.547, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.8864, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.794029951095581, | |
| "eval_runtime": 21.6648, | |
| "eval_samples_per_second": 453.039, | |
| "eval_steps_per_second": 3.554, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.8864, | |
| "step": 15340 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.7938315868377686, | |
| "eval_runtime": 21.6955, | |
| "eval_samples_per_second": 452.397, | |
| "eval_steps_per_second": 3.549, | |
| "step": 15340 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.8864, | |
| "step": 18408 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.7939543724060059, | |
| "eval_runtime": 21.6425, | |
| "eval_samples_per_second": 453.506, | |
| "eval_steps_per_second": 3.558, | |
| "step": 18408 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 1.8864, | |
| "step": 21476 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.7943979501724243, | |
| "eval_runtime": 21.6659, | |
| "eval_samples_per_second": 453.016, | |
| "eval_steps_per_second": 3.554, | |
| "step": 21476 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.8864, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.7938576936721802, | |
| "eval_runtime": 21.7538, | |
| "eval_samples_per_second": 451.185, | |
| "eval_steps_per_second": 3.54, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 1.8864, | |
| "step": 27612 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.7938728332519531, | |
| "eval_runtime": 21.8112, | |
| "eval_samples_per_second": 449.998, | |
| "eval_steps_per_second": 3.53, | |
| "step": 27612 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 1.8864, | |
| "step": 30680 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.7940329313278198, | |
| "eval_runtime": 21.9136, | |
| "eval_samples_per_second": 447.895, | |
| "eval_steps_per_second": 3.514, | |
| "step": 30680 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 30680, | |
| "total_flos": 1.1616577644396544e+17, | |
| "train_loss": 1.8864124283175114, | |
| "train_runtime": 22676.8207, | |
| "train_samples_per_second": 865.867, | |
| "train_steps_per_second": 6.765 | |
| } | |
| ], | |
| "max_steps": 153400, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.1616577644396544e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |