| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 1908, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.796187698841095, | |
| "learning_rate": 1.6677148846960168e-05, | |
| "loss": 0.8103, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6693548387096774, | |
| "eval_loss": 0.41790342330932617, | |
| "eval_runtime": 2.5216, | |
| "eval_samples_per_second": 1229.4, | |
| "eval_steps_per_second": 25.778, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.7648549675941467, | |
| "learning_rate": 1.3354297693920338e-05, | |
| "loss": 0.3223, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.834516129032258, | |
| "eval_loss": 0.15564045310020447, | |
| "eval_runtime": 2.6982, | |
| "eval_samples_per_second": 1148.921, | |
| "eval_steps_per_second": 24.09, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.9308156967163086, | |
| "learning_rate": 1.0031446540880504e-05, | |
| "loss": 0.1606, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8838709677419355, | |
| "eval_loss": 0.08456186950206757, | |
| "eval_runtime": 2.5849, | |
| "eval_samples_per_second": 1199.278, | |
| "eval_steps_per_second": 25.146, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.5661836266517639, | |
| "learning_rate": 6.708595387840672e-06, | |
| "loss": 0.1075, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9067741935483871, | |
| "eval_loss": 0.06164713203907013, | |
| "eval_runtime": 2.6514, | |
| "eval_samples_per_second": 1169.202, | |
| "eval_steps_per_second": 24.516, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.4375452995300293, | |
| "learning_rate": 3.385744234800839e-06, | |
| "loss": 0.0876, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.915483870967742, | |
| "eval_loss": 0.053106848150491714, | |
| "eval_runtime": 2.5807, | |
| "eval_samples_per_second": 1201.216, | |
| "eval_steps_per_second": 25.187, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.47591638565063477, | |
| "learning_rate": 6.289308176100629e-08, | |
| "loss": 0.0795, | |
| "step": 1902 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 1908, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 1000000000.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 495439677408900.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.4117223254186742, | |
| "num_train_epochs": 6, | |
| "temperature": 2 | |
| } | |
| } | |