You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardexpand all lines: text_chunker_app.py
+1-1
Original file line number
Diff line number
Diff line change
@@ -60,7 +60,7 @@ def process_text(text):
60
60
response_model=TextChunks,
61
61
messages=[
62
62
{"role": "system", "content": "You are an AI assistant tasked with chunking a text into cohesive sections. Your goal is to create chunks that maintain topic coherence and context."},
63
-
{"role": "user", "content": f"Here's a text with numbered artifacts. Determine the best chunks by specifying start and end artifact numbers. Make the chunks as large as possible while maintaining coherence. Provide a thorough context for each chunk, including information from the rest of the text to ensure the chunk makes good sense. Here's the text:\n\n{text_with_artifacts}"}
63
+
{"role": "user", "content": f"Here's a text with numbered artifacts. Determine the best chunks by specifying start and end artifact numbers. Make the chunks as large as possible while maintaining coherence. Provide a thorough context for each chunk, including information from the rest of the text to ensure the chunk makes good sense. Ensure no overlap between chunks, and also not gaps. the end of one chunk should be the start of the next. Here's the text:\n\n{text_with_artifacts}"}
0 commit comments