| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0001, | |
| "loss": 2.0446, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7752, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6076455696202532, | |
| "eval_loss": 1.8067126274108887, | |
| "eval_runtime": 5.0018, | |
| "eval_samples_per_second": 99.965, | |
| "eval_steps_per_second": 12.596, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 24.4, | |
| "eval_f1": 35.27913510466144, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7141, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6576, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6469, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.607746835443038, | |
| "eval_loss": 1.810449481010437, | |
| "eval_runtime": 4.5227, | |
| "eval_samples_per_second": 110.552, | |
| "eval_steps_per_second": 13.93, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 24.2, | |
| "eval_f1": 33.990575470281364, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4453, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0001, | |
| "loss": 1.474, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6055696202531645, | |
| "eval_loss": 1.8687865734100342, | |
| "eval_runtime": 4.7138, | |
| "eval_samples_per_second": 106.072, | |
| "eval_steps_per_second": 13.365, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 23.0, | |
| "eval_f1": 33.30356143856145, | |
| "step": 750 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 12500, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 4.789364497947034e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |