| { | |
| "best_metric": 0.863342821598053, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_mnli_256/checkpoint-27612", | |
| "epoch": 14.0, | |
| "global_step": 42952, | |
| "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": 1.0008, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5404992358634743, | |
| "eval_loss": 0.9489795565605164, | |
| "eval_runtime": 22.1002, | |
| "eval_samples_per_second": 444.113, | |
| "eval_steps_per_second": 3.484, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.9205, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5674987264391238, | |
| "eval_loss": 0.9166129231452942, | |
| "eval_runtime": 22.1405, | |
| "eval_samples_per_second": 443.305, | |
| "eval_steps_per_second": 3.478, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.8928, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.578604177279674, | |
| "eval_loss": 0.902202844619751, | |
| "eval_runtime": 22.2149, | |
| "eval_samples_per_second": 441.82, | |
| "eval_steps_per_second": 3.466, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.872, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5967396841569027, | |
| "eval_loss": 0.8842912316322327, | |
| "eval_runtime": 22.1304, | |
| "eval_samples_per_second": 443.507, | |
| "eval_steps_per_second": 3.479, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.8531, | |
| "step": 15340 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5959246051961283, | |
| "eval_loss": 0.8806653618812561, | |
| "eval_runtime": 22.1032, | |
| "eval_samples_per_second": 444.053, | |
| "eval_steps_per_second": 3.484, | |
| "step": 15340 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.8359, | |
| "step": 18408 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5998981151299032, | |
| "eval_loss": 0.8763103485107422, | |
| "eval_runtime": 22.114, | |
| "eval_samples_per_second": 443.837, | |
| "eval_steps_per_second": 3.482, | |
| "step": 18408 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.8197, | |
| "step": 21476 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6009169638308711, | |
| "eval_loss": 0.8814870119094849, | |
| "eval_runtime": 22.1114, | |
| "eval_samples_per_second": 443.889, | |
| "eval_steps_per_second": 3.482, | |
| "step": 21476 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.8028, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5933774834437087, | |
| "eval_loss": 0.9012145400047302, | |
| "eval_runtime": 22.1167, | |
| "eval_samples_per_second": 443.782, | |
| "eval_steps_per_second": 3.482, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.786, | |
| "step": 27612 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6190524707080999, | |
| "eval_loss": 0.863342821598053, | |
| "eval_runtime": 22.1033, | |
| "eval_samples_per_second": 444.052, | |
| "eval_steps_per_second": 3.484, | |
| "step": 27612 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.769, | |
| "step": 30680 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6097809475292919, | |
| "eval_loss": 0.8733872771263123, | |
| "eval_runtime": 22.175, | |
| "eval_samples_per_second": 442.616, | |
| "eval_steps_per_second": 3.472, | |
| "step": 30680 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.752, | |
| "step": 33748 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6220071319409067, | |
| "eval_loss": 0.8681850433349609, | |
| "eval_runtime": 22.0877, | |
| "eval_samples_per_second": 444.365, | |
| "eval_steps_per_second": 3.486, | |
| "step": 33748 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.736, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.617524197656648, | |
| "eval_loss": 0.8740953803062439, | |
| "eval_runtime": 22.1771, | |
| "eval_samples_per_second": 442.573, | |
| "eval_steps_per_second": 3.472, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.7204, | |
| "step": 39884 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6047885888945491, | |
| "eval_loss": 0.8993946313858032, | |
| "eval_runtime": 22.1027, | |
| "eval_samples_per_second": 444.063, | |
| "eval_steps_per_second": 3.484, | |
| "step": 39884 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.7038, | |
| "step": 42952 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6079470198675496, | |
| "eval_loss": 0.8940238356590271, | |
| "eval_runtime": 22.1071, | |
| "eval_samples_per_second": 443.976, | |
| "eval_steps_per_second": 3.483, | |
| "step": 42952 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "step": 42952, | |
| "total_flos": 1.4482128069931827e+17, | |
| "train_loss": 0.8189185247849343, | |
| "train_runtime": 37903.3591, | |
| "train_samples_per_second": 518.031, | |
| "train_steps_per_second": 4.047 | |
| } | |
| ], | |
| "max_steps": 153400, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.4482128069931827e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |