taf2/timeout-gpt-4preview

## timeout-gpt-4preview
sub_topics = [ ... ] # say you give it like 800 topics.

prompt =
Here is a set of JSON encoded topics: #{sub_topics.to_json}.
                   Please generalize these into a smaller set of no more then 10 topics.
                   Your new generalized topics may be expressed as multiple words.
                   Output as a JSON object of the new topics and the sub topics that belong to each topic.
        e.g. given ['sub topic 1', 'sub topic 2', 'sub topic 3', ...] group them into parent topics like {'topic1': ['sub topic 1', 'sub topic
        outputing as JSON

 system = You are an expert at generalizing topics.  You review lists of topics and group them into similar topics.
  Output as a JSON object of the new topics and the sub topics that belong to each topic.\n        e.g. given ['sub topic 1', 'sub topic 2', 'sub topic 3', ...] group them into parent topics like {'topic1': ['sub topic 1', 'sub topic 2'], 'topic2': ['sub topic 3', ...]}\n        outputing as JSON
	sub_topics = [ ... ] # say you give it like 800 topics.

	prompt =
	Here is a set of JSON encoded topics: #{sub_topics.to_json}.
	Please generalize these into a smaller set of no more then 10 topics.
	Your new generalized topics may be expressed as multiple words.
	Output as a JSON object of the new topics and the sub topics that belong to each topic.
	e.g. given ['sub topic 1', 'sub topic 2', 'sub topic 3', ...] group them into parent topics like {'topic1': ['sub topic 1', 'sub topic
	outputing as JSON

	system = You are an expert at generalizing topics. You review lists of topics and group them into similar topics.
	Output as a JSON object of the new topics and the sub topics that belong to each topic.\n e.g. given ['sub topic 1', 'sub topic 2', 'sub topic 3', ...] group them into parent topics like {'topic1': ['sub topic 1', 'sub topic 2'], 'topic2': ['sub topic 3', ...]}\n outputing as JSON