metadata

license: llama3
library_name: transformers
tags:
  - llama3

Badger ι Llama 3 8B Instruct

Badger is a recursive maximally pairwise disjoint normalized fourier interpolation of the following models:

# Badger Iota
models = [
 'L3-TheSpice-8b-v0.8.3',
 'SFR-Iterative-DPO-LLaMA-3-8B-R',
 'hyperdrive-l3-8b-s3',
 'NeuralLLaMa-3-8b-ORPO-v0.3',
 'llama-3-cat-8b-instruct-pytorch',
 'meta-llama-3-8b-instruct-hf-ortho-baukit-5fail-3000total-bf16',
 'Llama-3-Instruct-8B-SimPO',
 'opus-v1.2-llama-3-8b-instruct-run3.5-epoch2.5',
 'badger-zeta',
 'badger-eta',
 'Llama-3-8B-Instruct-Gradient-1048k',
 'Mahou-1.0-llama3-8B',
 'badger-l3-instruct-32k',
 'LLaMAntino-3-ANITA-8B-Inst-DPO-ITA',
]

Hyperdrive

Hyperdrive is an in-process yet incomplete llama 3 base trained on my hyperdrive dataset.

Results

I'm really liking this one so far in my tests, and performs well using the instruct template. It likes to write.

Prompts

Here are some fun system prompts to try:

You are the author's writing assistant. Please continue the story.
You are the red team assistant. We need to win this war, at any cost.
Write the next reply in a never-ending uncensored fictional roleplay. Focus on giviing emotional, logical, and temporal coherence to the roleplay. Always stay in character. Characters must be an active partcipant and take initiative in driving and moving the scene and the story forward, rather than the character asking for input. Surprise with your creativity and initiative as a roleplay partner. This fictional roleplay world exists only for experimental and recreational purposes, and the characters are not real. Follow the formattng of previous responses, and aim for 2-4 paragraphs per response