| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 8.999349381912817, |
| "global_step": 6912, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.65, |
| "learning_rate": 4.7287326388888894e-05, |
| "loss": 2.9288, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_bleu": 12.897750004409371, |
| "eval_loss": 2.277336835861206, |
| "eval_runtime": 3373.4947, |
| "eval_samples_per_second": 2.43, |
| "eval_steps_per_second": 0.304, |
| "step": 768 |
| }, |
| { |
| "epoch": 1.3, |
| "learning_rate": 4.457465277777778e-05, |
| "loss": 2.3415, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.95, |
| "learning_rate": 4.186197916666667e-05, |
| "loss": 2.1063, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_bleu": 15.060690733469018, |
| "eval_loss": 2.0984933376312256, |
| "eval_runtime": 3630.6371, |
| "eval_samples_per_second": 2.258, |
| "eval_steps_per_second": 0.282, |
| "step": 1536 |
| }, |
| { |
| "epoch": 2.6, |
| "learning_rate": 3.914930555555556e-05, |
| "loss": 1.884, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_bleu": 16.01923837036625, |
| "eval_loss": 2.0154550075531006, |
| "eval_runtime": 3598.1389, |
| "eval_samples_per_second": 2.278, |
| "eval_steps_per_second": 0.285, |
| "step": 2304 |
| }, |
| { |
| "epoch": 3.26, |
| "learning_rate": 3.643663194444444e-05, |
| "loss": 1.7932, |
| "step": 2500 |
| }, |
| { |
| "epoch": 3.91, |
| "learning_rate": 3.372395833333333e-05, |
| "loss": 1.6813, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_bleu": 16.69411603598328, |
| "eval_loss": 1.9759808778762817, |
| "eval_runtime": 3156.7439, |
| "eval_samples_per_second": 2.597, |
| "eval_steps_per_second": 0.325, |
| "step": 3072 |
| }, |
| { |
| "epoch": 4.56, |
| "learning_rate": 3.1011284722222224e-05, |
| "loss": 1.5634, |
| "step": 3500 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_bleu": 17.655854646910004, |
| "eval_loss": 1.9511423110961914, |
| "eval_runtime": 3562.0988, |
| "eval_samples_per_second": 2.301, |
| "eval_steps_per_second": 0.288, |
| "step": 3840 |
| }, |
| { |
| "epoch": 5.21, |
| "learning_rate": 2.8298611111111113e-05, |
| "loss": 1.5116, |
| "step": 4000 |
| }, |
| { |
| "epoch": 5.86, |
| "learning_rate": 2.55859375e-05, |
| "loss": 1.4307, |
| "step": 4500 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_bleu": 17.453809817741682, |
| "eval_loss": 1.9419958591461182, |
| "eval_runtime": 3674.8309, |
| "eval_samples_per_second": 2.231, |
| "eval_steps_per_second": 0.279, |
| "step": 4608 |
| }, |
| { |
| "epoch": 6.51, |
| "learning_rate": 2.287326388888889e-05, |
| "loss": 1.3663, |
| "step": 5000 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_bleu": 17.864149441439213, |
| "eval_loss": 1.9385316371917725, |
| "eval_runtime": 3083.4711, |
| "eval_samples_per_second": 2.658, |
| "eval_steps_per_second": 0.332, |
| "step": 5376 |
| }, |
| { |
| "epoch": 7.16, |
| "learning_rate": 2.016059027777778e-05, |
| "loss": 1.3295, |
| "step": 5500 |
| }, |
| { |
| "epoch": 7.81, |
| "learning_rate": 1.7447916666666666e-05, |
| "loss": 1.2783, |
| "step": 6000 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_bleu": 17.90688441590411, |
| "eval_loss": 1.9373716115951538, |
| "eval_runtime": 3414.7309, |
| "eval_samples_per_second": 2.4, |
| "eval_steps_per_second": 0.3, |
| "step": 6144 |
| }, |
| { |
| "epoch": 8.46, |
| "learning_rate": 1.4735243055555556e-05, |
| "loss": 1.2304, |
| "step": 6500 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_bleu": 18.254785590207636, |
| "eval_loss": 1.9363410472869873, |
| "eval_runtime": 3545.3415, |
| "eval_samples_per_second": 2.312, |
| "eval_steps_per_second": 0.289, |
| "step": 6912 |
| } |
| ], |
| "max_steps": 9216, |
| "num_train_epochs": 12, |
| "total_flos": 5739521784938496.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
