From 145a43a0000d4c502c0c8b3693fb04ff95aac072 Mon Sep 17 00:00:00 2001
From: Alephrin <84307744+Alephrin@users.noreply.github.com>
Date: Mon, 17 Jul 2023 04:53:47 -0600
Subject: [PATCH] Removed extra load_in_4bit.

---
 modeling/inference_models/generic_hf_torch/class.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/modeling/inference_models/generic_hf_torch/class.py b/modeling/inference_models/generic_hf_torch/class.py
index b51d8f66..9a59650e 100644
--- a/modeling/inference_models/generic_hf_torch/class.py
+++ b/modeling/inference_models/generic_hf_torch/class.py
@@ -80,7 +80,6 @@ class model_backend(HFTorchInferenceModel):
         if self.use_4_bit:
             self.lazy_load = False
             tf_kwargs.update({
-                "load_in_4bit": True,
                 "quantization_config":BitsAndBytesConfig(
                     load_in_4bit=True,
                     bnb_4bit_compute_dtype=torch.float16,