| { |
| "best_metric": 1.3110549449920654, |
| "best_model_checkpoint": "save/gpt2-canard2/checkpoint-2952", |
| "epoch": 8.998731286475515, |
| "global_step": 4428, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "eval_loss": 1.4947829246520996, |
| "eval_runtime": 41.6415, |
| "eval_samples_per_second": 82.37, |
| "step": 492 |
| }, |
| { |
| "epoch": 1.02, |
| "learning_rate": 3e-05, |
| "loss": 5.716, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 1.3761365413665771, |
| "eval_runtime": 41.645, |
| "eval_samples_per_second": 82.363, |
| "step": 984 |
| }, |
| { |
| "epoch": 2.03, |
| "learning_rate": 2.8394004282655247e-05, |
| "loss": 1.3967, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 1.3401412963867188, |
| "eval_runtime": 41.7345, |
| "eval_samples_per_second": 82.186, |
| "step": 1476 |
| }, |
| { |
| "epoch": 3.05, |
| "learning_rate": 2.678800856531049e-05, |
| "loss": 1.2189, |
| "step": 1500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 1.3260481357574463, |
| "eval_runtime": 41.812, |
| "eval_samples_per_second": 82.034, |
| "step": 1968 |
| }, |
| { |
| "epoch": 4.06, |
| "learning_rate": 2.518201284796574e-05, |
| "loss": 1.1113, |
| "step": 2000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 1.3199776411056519, |
| "eval_runtime": 41.8477, |
| "eval_samples_per_second": 81.964, |
| "step": 2460 |
| }, |
| { |
| "epoch": 5.08, |
| "learning_rate": 2.3576017130620987e-05, |
| "loss": 1.0324, |
| "step": 2500 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 1.3110549449920654, |
| "eval_runtime": 41.8913, |
| "eval_samples_per_second": 81.879, |
| "step": 2952 |
| }, |
| { |
| "epoch": 6.1, |
| "learning_rate": 2.197002141327623e-05, |
| "loss": 0.9693, |
| "step": 3000 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 1.321687936782837, |
| "eval_runtime": 41.8875, |
| "eval_samples_per_second": 81.886, |
| "step": 3444 |
| }, |
| { |
| "epoch": 7.11, |
| "learning_rate": 2.0364025695931477e-05, |
| "loss": 0.915, |
| "step": 3500 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 1.3250563144683838, |
| "eval_runtime": 41.9051, |
| "eval_samples_per_second": 81.852, |
| "step": 3936 |
| }, |
| { |
| "epoch": 8.13, |
| "learning_rate": 1.8758029978586724e-05, |
| "loss": 0.8714, |
| "step": 4000 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 1.3383173942565918, |
| "eval_runtime": 41.8731, |
| "eval_samples_per_second": 81.914, |
| "step": 4428 |
| }, |
| { |
| "epoch": 9.0, |
| "step": 4428, |
| "total_flos": 54234126510981120, |
| "train_runtime": 11078.034, |
| "train_samples_per_second": 0.888 |
| } |
| ], |
| "max_steps": 9840, |
| "num_train_epochs": 20, |
| "total_flos": 54234126510981120, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
