| { | |
| "best_metric": 0.2617154121398926, | |
| "best_model_checkpoint": "outputs/t5-base-detox/checkpoint-1080", | |
| "epoch": 8.995391705069125, | |
| "global_step": 1215, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9e-05, | |
| "loss": 1.337, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.48098286986351013, | |
| "eval_runtime": 18.2812, | |
| "eval_samples_per_second": 169.573, | |
| "eval_steps_per_second": 21.224, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5238, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.38857772946357727, | |
| "eval_runtime": 18.7764, | |
| "eval_samples_per_second": 165.101, | |
| "eval_steps_per_second": 20.664, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7e-05, | |
| "loss": 0.4301, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.33775028586387634, | |
| "eval_runtime": 18.505, | |
| "eval_samples_per_second": 167.522, | |
| "eval_steps_per_second": 20.967, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 6e-05, | |
| "loss": 0.3755, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.31217730045318604, | |
| "eval_runtime": 18.8887, | |
| "eval_samples_per_second": 164.119, | |
| "eval_steps_per_second": 20.541, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3359, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.2910080552101135, | |
| "eval_runtime": 18.2985, | |
| "eval_samples_per_second": 169.413, | |
| "eval_steps_per_second": 21.204, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3068, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.2737014591693878, | |
| "eval_runtime": 18.6417, | |
| "eval_samples_per_second": 166.294, | |
| "eval_steps_per_second": 20.814, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2861, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.2709713876247406, | |
| "eval_runtime": 18.2786, | |
| "eval_samples_per_second": 169.597, | |
| "eval_steps_per_second": 21.227, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2744, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.2617154121398926, | |
| "eval_runtime": 18.6478, | |
| "eval_samples_per_second": 166.24, | |
| "eval_steps_per_second": 20.807, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2649, | |
| "step": 1215 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.2630476653575897, | |
| "eval_runtime": 18.456, | |
| "eval_samples_per_second": 167.967, | |
| "eval_steps_per_second": 21.023, | |
| "step": 1215 | |
| } | |
| ], | |
| "max_steps": 1350, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.052005203836928e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |