Edit model card

Fi Nemo SFT

⚠️ This is an old model. It works, but it's certainly not amazing.

About

Fi Nemo is a roleplay model based on Mistral-Nemo-Instruct-2407. It is trained on a large amount of medium to high-quality fanfiction, with high-quality annotations.

Prompt Format

This model uses the same prompt format as Mistral-Nemo-Instruct-2407. You can prompt it with "Write..." instructions, however it may perform better if you simply tell it what actions happen in the text.

Tips and Tricks

  • The first message will set the tone for the next.
  • You may need to regenerate a few times, until you get the style and pacing that you want.
Downloads last month
66
GGUF
Model size
12.2B params
Architecture
llama

4-bit

16-bit

Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for hikitoxin/fi-nemo-sft-prelim

Quantized
(82)
this model

Dataset used to train hikitoxin/fi-nemo-sft-prelim