| { | |
| "best_metric": 0.8459745645523071, | |
| "best_model_checkpoint": "./Zeroshot/01-12-23-NousResearch-Nous-Hermes-Llama2-13b_multilang-dataset-3.0.3-portuguese-2_epochs-10_batch_2/checkpoints/checkpoint-3386", | |
| "epoch": 0.9998523549387273, | |
| "global_step": 3386, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 5.8948611931482576e-05, | |
| "loss": 1.3183, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00011801535735380982, | |
| "loss": 0.9343, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00017708210277613703, | |
| "loss": 0.9005, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00023614884819846425, | |
| "loss": 0.8859, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0002952155936207915, | |
| "loss": 0.8707, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00035416420555227406, | |
| "loss": 0.8523, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.8459745645523071, | |
| "eval_runtime": 133.0397, | |
| "eval_samples_per_second": 11.32, | |
| "eval_steps_per_second": 1.421, | |
| "step": 3386 | |
| } | |
| ], | |
| "max_steps": 33860, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.510076088815616e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |