configs | ||
figs | ||
peft@098962fa65 | ||
transformers@cae78c46d6 | ||
.gitignore | ||
.gitmodules | ||
clean.py | ||
data.py | ||
env.yaml | ||
generate.py | ||
read.py | ||
README.md | ||
requirements.txt | ||
train.py | ||
TRAINING_LOG.md |
GPT4All
Demo, data and code to train an assistant-style large language model
Try it yourself
-- TODO LLAMA C++ code
Reproducibility
You can find trained LoRa model weights at:
We are not distributing LLaMa 7B checkpoint they need to be used in association with.
To reproduce our LoRA training run, do the following:
Setup
Clone the repo
git clone --recurse-submodules git@github.com:nomic-ai/gpt4all.git
git submodule configure && git submodule update
Setup the environment
python -m pip install -r requirements.txt
cd transformers
pip install -e .
cd ../peft
pip install -e .
Generate
python generate.py --config configs/generate/generate.yaml --prompt "Write a script to reverse a string in Python
Train
accelerate launch --dynamo_backend=inductor --num_processes=8 --num_machines=1 --machine_rank=0 --deepspeed_multinode_launcher standard --mixed_precision=bf16 --use_deepspeed --deepspeed_config_file=configs/deepspeed/ds_config.json train.py --config configs/train/finetune-7b.yaml
If you utilize this reposistory, models or data in a downstream project, please consider citing it with:
@misc{gpt4all,
author = {Yuvanesh Anand and Zachary Nussbaum and Brandon Duderstadt and Benjamin Schmidt and Andriy Mulyar},
title = {GPT4All: Training an Assistant-style Chatbot with Large Scale Data Distillation from GPT-3.5-Turbo},
year = {2023},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/nomic-ai/gpt4all}},
}