| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 1420, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 8.591549295774647e-05, | |
| "loss": 0.9777, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.8169013857841492, | |
| "eval_loss": 0.6019795536994934, | |
| "eval_runtime": 117.4868, | |
| "eval_samples_per_second": 4.835, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 7.183098591549297e-05, | |
| "loss": 0.5253, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_accuracy": 0.8573943376541138, | |
| "eval_loss": 0.44053006172180176, | |
| "eval_runtime": 130.6116, | |
| "eval_samples_per_second": 4.349, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 5.774647887323944e-05, | |
| "loss": 0.3972, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_accuracy": 0.8802816867828369, | |
| "eval_loss": 0.3814992308616638, | |
| "eval_runtime": 131.0373, | |
| "eval_samples_per_second": 4.335, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 4.366197183098591e-05, | |
| "loss": 0.2095, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_accuracy": 0.8661971688270569, | |
| "eval_loss": 0.4881167411804199, | |
| "eval_runtime": 116.8147, | |
| "eval_samples_per_second": 4.862, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 2.9577464788732395e-05, | |
| "loss": 0.126, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "eval_accuracy": 0.9014084339141846, | |
| "eval_loss": 0.37247011065483093, | |
| "eval_runtime": 130.6973, | |
| "eval_samples_per_second": 4.346, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 1.556338028169014e-05, | |
| "loss": 0.0865, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "eval_accuracy": 0.8978873491287231, | |
| "eval_loss": 0.40177592635154724, | |
| "eval_runtime": 131.5193, | |
| "eval_samples_per_second": 4.319, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 1.4788732394366198e-06, | |
| "loss": 0.0579, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "eval_accuracy": 0.9031690359115601, | |
| "eval_loss": 0.3632720112800598, | |
| "eval_runtime": 116.9733, | |
| "eval_samples_per_second": 4.856, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 1420, | |
| "total_flos": 3.369279479928036e+16, | |
| "train_runtime": 6152.5452, | |
| "train_samples_per_second": 0.231 | |
| } | |
| ], | |
| "max_steps": 1420, | |
| "num_train_epochs": 5, | |
| "total_flos": 3.369279479928036e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |