diff --git a/modules/llm/Mistral-7b/Mistral-7b_fine_tune.py b/modules/llm/Mistral-7b/Mistral-7b_fine_tune.py
index a41fc61c30d68dd36a930fbe39e5d816fa49af10..99a152d8cc983ce001df03ca78db2c7473975489 100644
--- a/modules/llm/Mistral-7b/Mistral-7b_fine_tune.py
+++ b/modules/llm/Mistral-7b/Mistral-7b_fine_tune.py
@@ -58,7 +58,6 @@ def fine_tune(base_model, new_model):
     training_arguments = TrainingArguments(
         per_device_train_batch_size=2,
         gradient_accumulation_steps=1,
-        num_train_epochs=2,
         learning_rate=1e-4,
         logging_steps=2,
         optim="adamw_torch",
diff --git a/modules/llm/Mixtral-8x7b/Mixtral-8x7b_fine_tune.py b/modules/llm/Mixtral-8x7b/Mixtral-8x7b_fine_tune.py
index d9e8126498f942694ecf7215d6bb21bf9c9e35f9..9210d02fb9ba2fb2bee455bbc58020ddec48b752 100644
--- a/modules/llm/Mixtral-8x7b/Mixtral-8x7b_fine_tune.py
+++ b/modules/llm/Mixtral-8x7b/Mixtral-8x7b_fine_tune.py
@@ -72,7 +72,6 @@ def fine_tuned(base_model, new_model):
         args=TrainingArguments(
             per_device_train_batch_size=2,
             gradient_accumulation_steps=1,
-            num_train_epochs=2,
             learning_rate=1e-4,
             logging_steps=2,
             optim="adamw_torch",