Merge branch 'train' of github.com:nomic-ai/gpt4all into train

This commit is contained in:
Zach Nussbaum 2023-03-28 18:48:19 +00:00
commit df39811918

View File

@ -1,13 +1,30 @@
# gpt4all
<h1 align="center">GPT4All</h1>
<p align="center">Demo, data and code to train an assistant-style large language model</p>
# Try it yourself
-- TODO LLAMA C++ code
# Setup
# Reproducibility
You can find trained LoRa model weights at:
- gpt4all-lora https://huggingface.co/nomic-ai/gpt4all-lora
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
```
@ -29,3 +46,18 @@ pip install -e .
## 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}},
}
```