reduce new token length

indexer.py

@@ -15,7 +15,7 @@ def load_raw_dataset():
 
     df["combined"] = df["input"] + " " + df["output"]
 
-    df = df.sample(n=min(5000, len(df)), random_state=42).reset_index(drop=True)
+    df = df.sample(n=min(5000, len(df)), random_state=42)
 
     docs = [
         Document(
@@ -44,8 +44,8 @@ def get_llm(model_name):
         "text2text-generation",
         model=model,
         tokenizer=tokenizer,
-        max_new_tokens=1024,
-        temperature=0.7,
+        max_new_tokens=512,
+        temperature=1,
         do_sample=True,
     )