| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9954493742889647, | |
| "global_step": 7000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.7653286978508224e-05, | |
| "loss": 0.3321, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 8.593664169311523, | |
| "eval_runtime": 118.099, | |
| "eval_samples_per_second": 211.695, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.975189633375474e-05, | |
| "loss": 0.2726, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 9.972230911254883, | |
| "eval_runtime": 118.1942, | |
| "eval_samples_per_second": 211.525, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.185050568900126e-05, | |
| "loss": 0.2505, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": 11.511611938476562, | |
| "eval_runtime": 118.1309, | |
| "eval_samples_per_second": 211.638, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.394911504424779e-05, | |
| "loss": 0.2275, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 10.211225509643555, | |
| "eval_runtime": 118.2397, | |
| "eval_samples_per_second": 211.443, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.6047724399494312e-05, | |
| "loss": 0.2173, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 13.594472885131836, | |
| "eval_runtime": 118.3052, | |
| "eval_samples_per_second": 211.326, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 8.146333754740834e-06, | |
| "loss": 0.2038, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 17.407012939453125, | |
| "eval_runtime": 117.4057, | |
| "eval_samples_per_second": 212.945, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.449431099873578e-07, | |
| "loss": 0.1963, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 18.987485885620117, | |
| "eval_runtime": 117.6706, | |
| "eval_samples_per_second": 212.466, | |
| "step": 7000 | |
| } | |
| ], | |
| "max_steps": 7032, | |
| "num_train_epochs": 1, | |
| "total_flos": 75170067805473600, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |