| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6298815822625347, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.000975, | |
| "loss": 1.0238, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.884382426738739, | |
| "eval_runtime": 4.1887, | |
| "eval_samples_per_second": 238.736, | |
| "eval_steps_per_second": 29.842, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00095, | |
| "loss": 0.9971, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 0.8562089800834656, | |
| "eval_runtime": 4.1399, | |
| "eval_samples_per_second": 241.551, | |
| "eval_steps_per_second": 30.194, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.000925, | |
| "loss": 0.9522, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.8645691871643066, | |
| "eval_runtime": 4.1915, | |
| "eval_samples_per_second": 238.576, | |
| "eval_steps_per_second": 29.822, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.9143, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.8668216466903687, | |
| "eval_runtime": 4.1753, | |
| "eval_samples_per_second": 239.502, | |
| "eval_steps_per_second": 29.938, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.000875, | |
| "loss": 0.9151, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.8562004566192627, | |
| "eval_runtime": 4.2561, | |
| "eval_samples_per_second": 234.956, | |
| "eval_steps_per_second": 29.369, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00085, | |
| "loss": 0.9263, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 0.850217342376709, | |
| "eval_runtime": 4.1404, | |
| "eval_samples_per_second": 241.521, | |
| "eval_steps_per_second": 30.19, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.000825, | |
| "loss": 0.9102, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 0.8581109642982483, | |
| "eval_runtime": 4.1878, | |
| "eval_samples_per_second": 238.788, | |
| "eval_steps_per_second": 29.849, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0008, | |
| "loss": 0.8866, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.8308799862861633, | |
| "eval_runtime": 4.1108, | |
| "eval_samples_per_second": 243.259, | |
| "eval_steps_per_second": 30.407, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0007750000000000001, | |
| "loss": 0.911, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 0.801396369934082, | |
| "eval_runtime": 4.1725, | |
| "eval_samples_per_second": 239.665, | |
| "eval_steps_per_second": 29.958, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00075, | |
| "loss": 0.8536, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 0.8181760907173157, | |
| "eval_runtime": 4.1738, | |
| "eval_samples_per_second": 239.59, | |
| "eval_steps_per_second": 29.949, | |
| "step": 5000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 20000, | |
| "num_train_epochs": 3, | |
| "save_steps": 5000, | |
| "total_flos": 8851794099093504.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |