| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.9988400994200495, | |
| "eval_steps": 500, | |
| "global_step": 3771, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.779398762157383e-05, | |
| "loss": 1.8729, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.5922507047653198, | |
| "eval_runtime": 120.9305, | |
| "eval_samples_per_second": 111.386, | |
| "eval_steps_per_second": 6.963, | |
| "step": 754 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.5587975243147657e-05, | |
| "loss": 1.2962, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.3377541998231653e-05, | |
| "loss": 1.1483, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.519166111946106, | |
| "eval_runtime": 120.9596, | |
| "eval_samples_per_second": 111.36, | |
| "eval_steps_per_second": 6.961, | |
| "step": 1508 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 1.1171529619805484e-05, | |
| "loss": 0.9934, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.5429869890213013, | |
| "eval_runtime": 120.7206, | |
| "eval_samples_per_second": 111.58, | |
| "eval_steps_per_second": 6.975, | |
| "step": 2263 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 8.961096374889478e-06, | |
| "loss": 0.9204, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 6.750663129973475e-06, | |
| "loss": 0.857, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.5413910150527954, | |
| "eval_runtime": 121.094, | |
| "eval_samples_per_second": 111.236, | |
| "eval_steps_per_second": 6.953, | |
| "step": 3017 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 4.540229885057471e-06, | |
| "loss": 0.7687, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.5838693380355835, | |
| "eval_runtime": 122.1252, | |
| "eval_samples_per_second": 110.297, | |
| "eval_steps_per_second": 6.895, | |
| "step": 3771 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4524, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 9.460281739721472e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |