From 7276dca933626eef587ac0adf53ab8ea06c2e9ac Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 27 Sep 2024 20:26:36 -0700
Subject: [PATCH] Fix a typo

---
 modules/exllamav2_hf.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/modules/exllamav2_hf.py b/modules/exllamav2_hf.py
index febb2c64..96a89429 100644
--- a/modules/exllamav2_hf.py
+++ b/modules/exllamav2_hf.py
@@ -48,9 +48,9 @@ class Exllamav2HF(PreTrainedModel):
             split = [float(alloc) for alloc in shared.args.gpu_split.split(",")]
 
         if shared.args.enable_tp:
-            model.load_tp(split)
+            self.ex_model.load_tp(split)
         elif not shared.args.autosplit:
-            model.load(split)
+            self.ex_model.load(split)
 
         # Determine the correct cache type
         if shared.args.cache_8bit: