| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.8368794326241136, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 3.817966903073286e-05, | |
| "loss": 2.8326, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 4.2973, | |
| "eval_gen_len": 46.6443, | |
| "eval_loss": 2.453827381134033, | |
| "eval_runtime": 220.0489, | |
| "eval_samples_per_second": 2.849, | |
| "eval_steps_per_second": 0.359, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 2.6359338061465723e-05, | |
| "loss": 2.4294, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 5.3365, | |
| "eval_gen_len": 44.7368, | |
| "eval_loss": 2.3174281120300293, | |
| "eval_runtime": 206.4318, | |
| "eval_samples_per_second": 3.037, | |
| "eval_steps_per_second": 0.383, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.4539007092198581e-05, | |
| "loss": 2.2246, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 2.7186761229314422e-06, | |
| "loss": 2.0925, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 2115, | |
| "num_train_epochs": 3, | |
| "total_flos": 3127134057160704.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |