| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.502251125562782, | |
| "global_step": 9000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.8999499749874938e-05, | |
| "loss": 3.2627, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.7998999499749875e-05, | |
| "loss": 3.2043, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 3.152320146560669, | |
| "eval_runtime": 135.0377, | |
| "eval_samples_per_second": 15.677, | |
| "eval_steps_per_second": 2.614, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.6998499249624815e-05, | |
| "loss": 3.1872, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.599799899949975e-05, | |
| "loss": 3.1659, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.1159095764160156, | |
| "eval_runtime": 134.9774, | |
| "eval_samples_per_second": 15.684, | |
| "eval_steps_per_second": 2.615, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.4997498749374688e-05, | |
| "loss": 3.0149, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.3996998499249626e-05, | |
| "loss": 3.0163, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 3.104168176651001, | |
| "eval_runtime": 135.046, | |
| "eval_samples_per_second": 15.676, | |
| "eval_steps_per_second": 2.614, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 1.2996498249124563e-05, | |
| "loss": 3.0252, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.19959979989995e-05, | |
| "loss": 3.0156, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 3.0884268283843994, | |
| "eval_runtime": 135.2307, | |
| "eval_samples_per_second": 15.655, | |
| "eval_steps_per_second": 2.61, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.0995497748874438e-05, | |
| "loss": 2.9118, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 9.994997498749376e-06, | |
| "loss": 2.9021, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 3.094877004623413, | |
| "eval_runtime": 134.8986, | |
| "eval_samples_per_second": 15.693, | |
| "eval_steps_per_second": 2.617, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 8.994497248624313e-06, | |
| "loss": 2.8996, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7.993996998499251e-06, | |
| "loss": 2.9048, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.087371587753296, | |
| "eval_runtime": 135.1793, | |
| "eval_samples_per_second": 15.661, | |
| "eval_steps_per_second": 2.611, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 6.9934967483741875e-06, | |
| "loss": 2.8394, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 5.992996498249125e-06, | |
| "loss": 2.819, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "eval_loss": 3.090629816055298, | |
| "eval_runtime": 133.0042, | |
| "eval_samples_per_second": 15.917, | |
| "eval_steps_per_second": 2.654, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 4.992496248124062e-06, | |
| "loss": 2.8219, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.991995997999e-06, | |
| "loss": 2.8267, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.085977077484131, | |
| "eval_runtime": 133.3062, | |
| "eval_samples_per_second": 15.881, | |
| "eval_steps_per_second": 2.648, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 2.991495747873937e-06, | |
| "loss": 2.776, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 1.9909954977488747e-06, | |
| "loss": 2.7743, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "eval_loss": 3.094261646270752, | |
| "eval_runtime": 133.0331, | |
| "eval_samples_per_second": 15.913, | |
| "eval_steps_per_second": 2.653, | |
| "step": 9000 | |
| } | |
| ], | |
| "max_steps": 9995, | |
| "num_train_epochs": 5, | |
| "total_flos": 2.546671435776e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |