Sanity check for fiction cleanup using mistral 7b | Learn AI Together | Page 1

#

Hello.
Could someone that has a bit of time give me a sanity check?

Problem: Legitimate fiction downloads(ao3 and so on) contains author's notes which I don't want to listen to using tts(used because of problems with my eyes).

Solution: Use python Ollama to feed mistral 7b paragraphs and ask to classify paragraphs into story or author's notes.

Question: Any better ideas of how this could be accomplished? Better models to use(laptop) or whether i should try to use an API instead?

Prompt:
You are a classifier. Read the following text and answer with ONLY ONE WORD:
"Yes" if the text is part of a fictional story, or "No" if it is an author's note or commentary.
Assume "No"s are found grouped at the beginning and end of the text. This is part {i} of {n}.
Text:
{text}
Answer one word: Yes or No

#

Sanity check for fiction cleanup

#

Sanity check for fiction cleanup using mistral 7b

#Sanity check for fiction cleanup using mistral 7b