Instructions to use bbunzeck/gpt-wee-large-curriculum with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use bbunzeck/gpt-wee-large-curriculum with Transformers:

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

pipe = pipeline("text-generation", model="bbunzeck/gpt-wee-large-curriculum")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("bbunzeck/gpt-wee-large-curriculum")
model = AutoModelForCausalLM.from_pretrained("bbunzeck/gpt-wee-large-curriculum")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use bbunzeck/gpt-wee-large-curriculum with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/bbunzeck/gpt-wee-large-curriculum

SGLang

How to use bbunzeck/gpt-wee-large-curriculum 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 "bbunzeck/gpt-wee-large-curriculum" \
    --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": "bbunzeck/gpt-wee-large-curriculum",
		"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 "bbunzeck/gpt-wee-large-curriculum" \
        --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": "bbunzeck/gpt-wee-large-curriculum",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use bbunzeck/gpt-wee-large-curriculum with Docker Model Runner:
```
docker model run hf.co/bbunzeck/gpt-wee-large-curriculum
```

GPT-2 model trained as ablation by team CLAUSE Bielefeld for the BabyLM challenge 2023
for more info, please check out the paper

Citation:

@inproceedings{bunzeck-zarriess-2023-gpt,
    title = "{GPT}-wee: How Small Can a Small Language Model Really Get?",
    author = "Bunzeck, Bastian  and
      Zarrie{\ss}, Sina",
    editor = "Warstadt, Alex  and
      Mueller, Aaron  and
      Choshen, Leshem  and
      Wilcox, Ethan  and
      Zhuang, Chengxu  and
      Ciro, Juan  and
      Mosquera, Rafael  and
      Paranjabe, Bhargavi  and
      Williams, Adina  and
      Linzen, Tal  and
      Cotterell, Ryan",
    booktitle = "Proceedings of the BabyLM Challenge at the 27th Conference on Computational Natural Language Learning",
    month = dec,
    year = "2023",
    address = "Singapore",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2023.conll-babylm.2",
    doi = "10.18653/v1/2023.conll-babylm.2",
    pages = "35--46",
}

Downloads last month: 1

Dataset used to train bbunzeck/gpt-wee-large-curriculum

Collection including bbunzeck/gpt-wee-large-curriculum

GPT-wee: How Small Can a Small Language Model Really Get?

Collection

Models trained as submission for BabyLM challenge 2023. Paper: https://aclanthology.org/2023.conll-babylm.2 • 6 items • Updated Nov 7, 2024