| { | |
| "best_metric": 0.22521649301052094, | |
| "best_model_checkpoint": "outputs/t5-v1_1-base-detox/checkpoint-1080", | |
| "epoch": 9.995391705069125, | |
| "global_step": 1350, | |
| "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": 8.9719, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.031940221786499, | |
| "eval_runtime": 22.7196, | |
| "eval_samples_per_second": 136.446, | |
| "eval_steps_per_second": 17.078, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8e-05, | |
| "loss": 1.9438, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6424035429954529, | |
| "eval_runtime": 22.8507, | |
| "eval_samples_per_second": 135.663, | |
| "eval_steps_per_second": 16.98, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7e-05, | |
| "loss": 0.8941, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.43842175602912903, | |
| "eval_runtime": 22.8845, | |
| "eval_samples_per_second": 135.463, | |
| "eval_steps_per_second": 16.955, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6694, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.3455743193626404, | |
| "eval_runtime": 22.9242, | |
| "eval_samples_per_second": 135.228, | |
| "eval_steps_per_second": 16.925, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5336, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.2819174826145172, | |
| "eval_runtime": 23.1392, | |
| "eval_samples_per_second": 133.972, | |
| "eval_steps_per_second": 16.768, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4519, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.24306097626686096, | |
| "eval_runtime": 22.7762, | |
| "eval_samples_per_second": 136.107, | |
| "eval_steps_per_second": 17.035, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4006, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.23186618089675903, | |
| "eval_runtime": 23.1423, | |
| "eval_samples_per_second": 133.954, | |
| "eval_steps_per_second": 16.766, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3771, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.22521649301052094, | |
| "eval_runtime": 22.8858, | |
| "eval_samples_per_second": 135.455, | |
| "eval_steps_per_second": 16.954, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3646, | |
| "step": 1215 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.22711366415023804, | |
| "eval_runtime": 23.0199, | |
| "eval_samples_per_second": 134.666, | |
| "eval_steps_per_second": 16.855, | |
| "step": 1215 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.3578, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.22696341574192047, | |
| "eval_runtime": 22.8653, | |
| "eval_samples_per_second": 135.576, | |
| "eval_steps_per_second": 16.969, | |
| "step": 1350 | |
| } | |
| ], | |
| "max_steps": 1350, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.3138971617968128e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |