---
license: wtfpl
---

# dril-instruct

putting [dril](https://twitter.com/dril) in a jar for future generations.

![picture of dril in a jar](https://github.com/lun-4/dril-instruct/blob/mistress/files/driljar.jpg?raw=true)

## Model Details

### Model Description

a LoRA finetune on top of [vicuna-13b-cocktail](https://huggingface.co/reeducator/vicuna-13b-cocktail),
made to act like dril's tweets when given an instruction to "Create a joke about X", where X is anything.

- **Developed by:** [lun-4](https://l4.pm), [dither](https://github.com/dithercat)
- **Model type:** LoRA
- **Language(s) (NLP):** English
- **License:** WTFPL
- **Finetuned from model:** [vicuna-13b-cocktail](https://huggingface.co/reeducator/vicuna-13b-cocktail)

### Model Sources [optional]

- **Repository:** https://github.com/lun-4/dril-instruct
- **Blogpost:** https://l4.pm/wiki/Personal%20Wiki/AI%20stuff/dril-instruct.html
- **Demo**: Nope

## Uses

by shitposters, for shitposting, this isn't useful for any other purpose

### Out-of-Scope Use

literally anything other than shitposting

## Bias, Risks, and Limitations

this model was finetuned on cocktail, which was made from [vicuna](https://lmsys.org/blog/2023-03-30-vicuna/)
but without the "ethical guardrails" AKA "As an AI language model, I can't" responses

## How to Get Started with the Model

1. get [text-generation-webui](https://github.com/oobabooga/text-generation-webui)
2. get the base model, [vicuna-13b-cocktail](https://huggingface.co/reeducator/vicuna-13b-cocktail) in the models folder
3. put this in the loras folder
4. there is a lot of hacking that i had to do to make loras work with GPTQ quantized models on my machine. those hacks are not portable
5. use the "Create a joke about X" model template.

## Training Details

### Training Data

around 3K dril tweets (that's what snscrape could get, even though there's 12K reported by twitter), and
some 10 or so hand-made instructions to the dril tweets

### Training Procedure

see blogpost

#### Preprocessing [optional]

see blogpost pls

#### Training Hyperparameters

- **Training regime:** int8

## Evaluation

see blogpost pls

## Environmental Impact

fuck if i know

<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->

Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).

- **Hardware Type:** NVidia A100 80GB
- **Hours used:** 1, not including 4 hours of trying to bang a training script together
- **Cloud Provider:** [RunPod](https://runpod.io)
- **Compute Region:** EU-Norway
- **Carbon Emitted:** 0.1 kg CO2 eq. (0.47 if you include the "banging rocks together" step)