LocalLLaMA

Most models are going to require some heavy prompting to get them even close to the prose you're looking for.

I've heard good things about Nous-Capybara-34B and Goliath 120b

This post is an automated archive from a submission made on /r/LocalLLaMA, powered by Fediverser software running on alien.top. Responses to this submission will not be seen by the original author until they claim ownership of their alien.top account. Please consider reaching out to them let them know about this post and help them migrate to Lemmy.

Lemmy users: you are still very much encouraged to participate in the discussion. There are still many other subscribers on !localllama@poweruser.forum that can benefit from your contribution and join in the conversation.

Reddit users: you can also join the fediverse right away by getting by visiting https://portal.alien.top. If you are looking for a Reddit alternative made for and by an independent community, check out Fediverser.

I could write a book about this.

First no model will be able to write well in style x and then in style y. Not at the size we can work with. Not at the size of ChatGPT either, although it would generally work better (it usually well approximates what the style is and falls into a Hollywood imagination of that style)

With finetuning you can make the model do one of those things.

## Instruction:

Rewrite the following text as a poem: Lister sat near the fireplace, looking at his arms which were covered in ointment. I find it hard to believe that pain can form words. Is this something internal or external? On his left arm, the message read "Dying" while on his right arm, it said "Living". He did not know what to think. A man who had been through a lot.

​

## Response:

Lister sat by the hearth and gazed

Upon the ointment-soaked clay

That coated his sinewy bare arms.

How could such agony frame

Such mystic characters? On one arm

The word "Dying" was graven clear;

On the other, "Living" appeared.

What could it mean? Ah! there,

A man whose life had been one tale

Of woes, of tortures, of crosses, of shame,

Who knew not what to hail as his own name.

Noromaid 20b and dolphin-mistral 7b do fairly well. Neither have much of a context length though.

Doing a full fine tune on Mistral 7B is the only way I’ve gotten human, literary text out of any of these models. Occasionally the vanilla Llama-2 70B will output something great. Yi-34B-Chat, while not by default a literary writer (it’s got that clunky, purple prose, GPT-4 feel to it) impressed me with its ability to write in a requested style.

The old Guanaco 33B and 65B models produced nice prose, but unfortunately they’re only 2048 context and they weren’t the best at following instructions.

As I understand LLMs basically write the average pattern of a billion books, so when you add gpt-4 and 3.5 data into the mix, which averages the average, things get boring very fast. For model suggestion, Yi-34b based ones look fine for literary purposes.

I think being very specific and editing (co-writing with the model) could help. Some LoRA training on specific books could be helpful to mimic a certain style.

High temperature and repetition penalty could help too.

Here are my top faves atm, not perfect but sometimes they output some decent content:

https://www.reddit.com/r/LocalLLaMA/comments/181w2b1/what_is_your_top_3_fav_llms_with_settings_for/