From e1d549bcd383f39bbeb5a72f33f54245c6711a63 Mon Sep 17 00:00:00 2001 From: Goekdeniz-Guelmez Date: Wed, 22 Jan 2025 21:03:21 +0100 Subject: [PATCH] nits --- llms/mlx_lm/tuner/dpo_trainer.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/llms/mlx_lm/tuner/dpo_trainer.py b/llms/mlx_lm/tuner/dpo_trainer.py index d58cfb8d..142f8aea 100644 --- a/llms/mlx_lm/tuner/dpo_trainer.py +++ b/llms/mlx_lm/tuner/dpo_trainer.py @@ -371,7 +371,7 @@ def train_dpo( if rank == 0: print( f"Iter {it}: " - f"Val loss {val_loss:.3f}, " + f"Val loss {val_loss:.8f}, " f"Val chosen reward {val_rewards[0]:.3f}, " f"Val rejected reward {val_rewards[1]:.3f}, " f"Val took {val_time:.3f}s", @@ -414,7 +414,7 @@ def train_dpo( if rank == 0: print( - f"Iter {it}: Train loss {train_loss:.3f}, " + f"Iter {it}: Train loss {train_loss:.8f}, " f"Chosen reward {train_rewards[0]:.3f}, " f"Rejected reward {train_rewards[1]:.3f}, " f"Learning Rate {learning_rate:.3e}, "