Self-Alignment with Instruction Backtranslation
Paper • 2308.06259 • Published • 43
How to use from
SGLangUse Docker images
docker run --gpus all \
--shm-size 32g \
-p 30000:30000 \
-v ~/.cache/huggingface:/root/.cache/huggingface \
--env "HF_TOKEN=<secret>" \
--ipc=host \
lmsysorg/sglang:latest \
python3 -m sglang.launch_server \
--model-path "Spico/Humback-M0" \
--host 0.0.0.0 \
--port 30000# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Spico/Humback-M0",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'Quick Links
🐋 Humback
The proposed Humback is a novel framework that can augment the instruction data for supervised fine-tuning with high quality.
This is a SFT (supervised fine-tuning) model $M_{0}$ for Humback reproduction.
This model is trained on the seed data.
The seed data is a sampled dataset from oasst1.
You may find more details and usage examples in Spico197/Humback .
📜 Reference
@misc{li2023selfalignment,
title={Self-Alignment with Instruction Backtranslation},
author={Xian Li and Ping Yu and Chunting Zhou and Timo Schick and Luke Zettlemoyer and Omer Levy and Jason Weston and Mike Lewis},
year={2023},
eprint={2308.06259},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
- Downloads last month
- 17
Install from pip and serve model
# Install SGLang from pip: pip install sglang# Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Spico/Humback-M0" \ --host 0.0.0.0 \ --port 30000# Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Spico/Humback-M0", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'