|
--- |
|
datasets: |
|
- gozfarb/ShareGPT_Vicuna_unfiltered |
|
--- |
|
# Convert tools |
|
https://github.com/practicaldreamer/vicuna_to_alpaca |
|
|
|
# Training tool |
|
https://github.com/oobabooga/text-generation-webui |
|
|
|
ATM I'm using 2023.05.04v0 of the dataset and training full context. |
|
|
|
# Notes: |
|
So im only training 1 epoch as full context 30b takes a long time to train. |
|
My 1 epoch will take me 8 days lol but lucly the LoRA feels fully functinal at epoch 1 as shown on my 13b one. |
|
|
|
# How to test? |
|
1. Download LLaMA-30B-HF: https://huggingface.co/Neko-Institute-of-Science/LLaMA-30B-HF |
|
2. Replace special_tokens_map.json and tokenizer_config.json using the ones on this repo. |
|
3. Rename LLaMA-30B-HF to vicuna-30b |
|
4. Load ooba: ```python server.py --listen --model vicuna-30b --load-in-8bit --chat --lora checkpoint-xxxx``` |
|
5. Instruct mode: Vicuna-v1 it will load Vicuna-v0 by defualt |
|
|
|
|
|
# Want to see it Training? |
|
https://wandb.ai/neko-science/VicUnLocked/runs/vx8yzwi7 |
