diff --git a/prompting/tasks/sentiment.py b/prompting/tasks/sentiment.py
index 47b8b1e73..7b4b58945 100644
--- a/prompting/tasks/sentiment.py
+++ b/prompting/tasks/sentiment.py
@@ -25,7 +25,6 @@ class SentimentAnalysisTask(Task):
 
     def __init__(self, llm_pipeline, context, create_reference=True):
         self.context = context
-
         self.query_prompt = QUERY_PROMPT_TEMPLATE.format(context=context.content)
         self.query = self.generate_query(llm_pipeline)
         self.reference = context.subtopic
diff --git a/prompting/tools/datasets/review.py b/prompting/tools/datasets/review.py
index 5c9300d91..41e512971 100644
--- a/prompting/tools/datasets/review.py
+++ b/prompting/tools/datasets/review.py
@@ -12,7 +12,7 @@ class ReviewDataset(TemplateDataset):
     SENTIMENTS = ["positive", "neutral", "negative"]
     # TODO: filter nonsense combinations of params
     # "Create a {topic} review of a {title} in the style of {mood} person in a {subtopic} tone. The review must be of {sentiment} sentiment."
-    query_template = "Create a {topic} review of a {title}. The review must be of {sentiment} sentiment."
+    query_template = "Create a {topic} review of a {title}. The review must be of {subtopic} sentiment."
     params = dict(
         topic=[
             "short",
@@ -44,5 +44,6 @@ class ReviewDataset(TemplateDataset):
             "company",
             "live event",
         ],
-        sentiment=SENTIMENTS,
-    )
\ No newline at end of file
+        subtopic=SENTIMENTS,
+    )
+    
\ No newline at end of file