| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 9.922822491730981, |
| "eval_steps": 500, |
| "global_step": 9000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.55, |
| "learning_rate": 2.834619625137817e-06, |
| "loss": 0.2705, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.748898678414097, |
| "eval_loss": 0.18952614068984985, |
| "eval_runtime": 5.6395, |
| "eval_samples_per_second": 40.252, |
| "eval_steps_per_second": 1.773, |
| "step": 907 |
| }, |
| { |
| "epoch": 1.1, |
| "learning_rate": 2.669239250275634e-06, |
| "loss": 0.2029, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.65, |
| "learning_rate": 2.503858875413451e-06, |
| "loss": 0.1943, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8193832599118943, |
| "eval_loss": 0.21802660822868347, |
| "eval_runtime": 5.6792, |
| "eval_samples_per_second": 39.97, |
| "eval_steps_per_second": 1.761, |
| "step": 1814 |
| }, |
| { |
| "epoch": 2.21, |
| "learning_rate": 2.338478500551268e-06, |
| "loss": 0.1796, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.76, |
| "learning_rate": 2.173098125689085e-06, |
| "loss": 0.1822, |
| "step": 2500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8370044052863436, |
| "eval_loss": 0.21766544878482819, |
| "eval_runtime": 5.7033, |
| "eval_samples_per_second": 39.801, |
| "eval_steps_per_second": 1.753, |
| "step": 2721 |
| }, |
| { |
| "epoch": 3.31, |
| "learning_rate": 2.0077177508269016e-06, |
| "loss": 0.1792, |
| "step": 3000 |
| }, |
| { |
| "epoch": 3.86, |
| "learning_rate": 1.842337375964719e-06, |
| "loss": 0.1804, |
| "step": 3500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8281938325991189, |
| "eval_loss": 0.23181109130382538, |
| "eval_runtime": 5.5689, |
| "eval_samples_per_second": 40.762, |
| "eval_steps_per_second": 1.796, |
| "step": 3628 |
| }, |
| { |
| "epoch": 4.41, |
| "learning_rate": 1.6769570011025359e-06, |
| "loss": 0.1674, |
| "step": 4000 |
| }, |
| { |
| "epoch": 4.96, |
| "learning_rate": 1.511576626240353e-06, |
| "loss": 0.1777, |
| "step": 4500 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8546255506607929, |
| "eval_loss": 0.19964338839054108, |
| "eval_runtime": 5.6954, |
| "eval_samples_per_second": 39.857, |
| "eval_steps_per_second": 1.756, |
| "step": 4535 |
| }, |
| { |
| "epoch": 5.51, |
| "learning_rate": 1.34619625137817e-06, |
| "loss": 0.1704, |
| "step": 5000 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8414096916299558, |
| "eval_loss": 0.21843558549880981, |
| "eval_runtime": 5.7142, |
| "eval_samples_per_second": 39.725, |
| "eval_steps_per_second": 1.75, |
| "step": 5442 |
| }, |
| { |
| "epoch": 6.06, |
| "learning_rate": 1.1808158765159868e-06, |
| "loss": 0.1762, |
| "step": 5500 |
| }, |
| { |
| "epoch": 6.62, |
| "learning_rate": 1.0154355016538037e-06, |
| "loss": 0.18, |
| "step": 6000 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8546255506607929, |
| "eval_loss": 0.22099512815475464, |
| "eval_runtime": 5.7111, |
| "eval_samples_per_second": 39.747, |
| "eval_steps_per_second": 1.751, |
| "step": 6349 |
| }, |
| { |
| "epoch": 7.17, |
| "learning_rate": 8.500551267916208e-07, |
| "loss": 0.1643, |
| "step": 6500 |
| }, |
| { |
| "epoch": 7.72, |
| "learning_rate": 6.846747519294377e-07, |
| "loss": 0.1708, |
| "step": 7000 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8502202643171806, |
| "eval_loss": 0.21724465489387512, |
| "eval_runtime": 5.5584, |
| "eval_samples_per_second": 40.839, |
| "eval_steps_per_second": 1.799, |
| "step": 7256 |
| }, |
| { |
| "epoch": 8.27, |
| "learning_rate": 5.192943770672547e-07, |
| "loss": 0.1822, |
| "step": 7500 |
| }, |
| { |
| "epoch": 8.82, |
| "learning_rate": 3.539140022050717e-07, |
| "loss": 0.1701, |
| "step": 8000 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8502202643171806, |
| "eval_loss": 0.22301247715950012, |
| "eval_runtime": 5.5557, |
| "eval_samples_per_second": 40.859, |
| "eval_steps_per_second": 1.8, |
| "step": 8163 |
| }, |
| { |
| "epoch": 9.37, |
| "learning_rate": 1.8853362734288866e-07, |
| "loss": 0.1705, |
| "step": 8500 |
| }, |
| { |
| "epoch": 9.92, |
| "learning_rate": 2.3153252480705623e-08, |
| "loss": 0.1762, |
| "step": 9000 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 9070, |
| "num_train_epochs": 10, |
| "save_steps": 1500, |
| "total_flos": 1535651327592000.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
