Instructions to use Data-Selection/PDS-470M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Data-Selection/PDS-470M with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Data-Selection/PDS-470M")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Data-Selection/PDS-470M")
model = AutoModelForCausalLM.from_pretrained("Data-Selection/PDS-470M")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Data-Selection/PDS-470M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Data-Selection/PDS-470M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Data-Selection/PDS-470M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Data-Selection/PDS-470M

SGLang

How to use Data-Selection/PDS-470M with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Data-Selection/PDS-470M" \
    --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": "Data-Selection/PDS-470M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use 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 "Data-Selection/PDS-470M" \
        --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": "Data-Selection/PDS-470M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Data-Selection/PDS-470M with Docker Model Runner:
```
docker model run hf.co/Data-Selection/PDS-470M
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

PDS-470M

paper | code | project page

PDS-470M is a 470M parameter Mistral architecture model pretrained from scratch using the PDS framework on data selected from the CC split of Redpajama.

The PDS framework is based on the Pontryagin's maximum principle for optimal pre-training data selection, offering strong theoretical support and scalability for training large language models.

Please refer to our paper for more details.

Overview of the theory:

Overview of the PDS framework:

Evaluation

PDS-selected data improves the performance of language models pre-trained from scratch and saves pre-training comptation. The improvement scales up to large model sizes.

Baseline

Conventional Pre-training

Citation

@article{gu2024data,
  title={Data Selection via Optimal Control for Language Models},
  author={Gu, Yuxian and Dong, Li and Wang, Hongning and Hao, Yaru and Dong, Qingxiu and Wei, Furu and Huang, Minlie},
  journal={arXiv preprint arXiv:2410.07064},
  year={2024}
}