| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 2430, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.742798353909465e-05, | |
| "loss": 1.947, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.48559670781893e-05, | |
| "loss": 1.8283, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 25.4528, | |
| "eval_gen_len": 11.9922, | |
| "eval_loss": 1.805336356163025, | |
| "eval_runtime": 261.4963, | |
| "eval_samples_per_second": 7.874, | |
| "step": 1215 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.2283950617283955e-05, | |
| "loss": 1.5834, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.971193415637861e-05, | |
| "loss": 1.4564, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 25.8034, | |
| "eval_gen_len": 11.933, | |
| "eval_loss": 1.820918083190918, | |
| "eval_runtime": 249.2526, | |
| "eval_samples_per_second": 8.261, | |
| "step": 2430 | |
| } | |
| ], | |
| "max_steps": 9720, | |
| "num_train_epochs": 8, | |
| "total_flos": 3.0060578130591744e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |