From 8a272cce6ca3c79517dbc6e52542d1971538632c Mon Sep 17 00:00:00 2001
From: Michael Noukhovitch <mnoukhov@gmail.com>
Date: Fri, 27 Oct 2023 13:53:13 -0400
Subject: [PATCH] hotfix for dpo trainer

addresses #914
---
 trl/trainer/dpo_trainer.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/trl/trainer/dpo_trainer.py b/trl/trainer/dpo_trainer.py
index 8ef7ffb431..60a4aeffe4 100644
--- a/trl/trainer/dpo_trainer.py
+++ b/trl/trainer/dpo_trainer.py
@@ -638,6 +638,7 @@ def evaluation_loop(
                     )
                 }
             )
+            self.state.log_history.pop()
 
         # Base evaluation
         initial_output = super().evaluation_loop(