| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9998104265402845, | |
| "global_step": 7911, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.683984325622551e-05, | |
| "loss": 0.7763, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.367968651245102e-05, | |
| "loss": 0.4148, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.051952976867653e-05, | |
| "loss": 0.3387, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3.7359373024902036e-05, | |
| "loss": 0.2855, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.4199216281127545e-05, | |
| "loss": 0.2469, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3.1039059537353053e-05, | |
| "loss": 0.2117, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2.7878902793578565e-05, | |
| "loss": 0.1833, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 2.471874604980407e-05, | |
| "loss": 0.1696, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 2.1558589306029582e-05, | |
| "loss": 0.1554, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.8398432562255087e-05, | |
| "loss": 0.1489, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1.5238275818480596e-05, | |
| "loss": 0.1319, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.2078119074706106e-05, | |
| "loss": 0.1197, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 8.917962330931615e-06, | |
| "loss": 0.1158, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 5.757805587157123e-06, | |
| "loss": 0.1105, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 2.597648843382632e-06, | |
| "loss": 0.1057, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 7911, | |
| "total_flos": 2.962265039645184e+17, | |
| "train_loss": 0.22759322628271503, | |
| "train_runtime": 16489.6196, | |
| "train_samples_per_second": 30.708, | |
| "train_steps_per_second": 0.48 | |
| } | |
| ], | |
| "max_steps": 7911, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.962265039645184e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |