Instructions to use EleutherAI/polyglot-ko-1.3b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use EleutherAI/polyglot-ko-1.3b with Transformers:

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

pipe = pipeline("text-generation", model="EleutherAI/polyglot-ko-1.3b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("EleutherAI/polyglot-ko-1.3b")
model = AutoModelForCausalLM.from_pretrained("EleutherAI/polyglot-ko-1.3b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use EleutherAI/polyglot-ko-1.3b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "EleutherAI/polyglot-ko-1.3b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EleutherAI/polyglot-ko-1.3b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/EleutherAI/polyglot-ko-1.3b

SGLang

How to use EleutherAI/polyglot-ko-1.3b 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 "EleutherAI/polyglot-ko-1.3b" \
    --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": "EleutherAI/polyglot-ko-1.3b",
		"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 "EleutherAI/polyglot-ko-1.3b" \
        --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": "EleutherAI/polyglot-ko-1.3b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use EleutherAI/polyglot-ko-1.3b with Docker Model Runner:
```
docker model run hf.co/EleutherAI/polyglot-ko-1.3b
```

I can't reproduce the kobest_hellaswag

by Woncheol - opened Jun 19, 2023

Discussion

Woncheol

Jun 19, 2023

As I tried to reproduce the polyglot evaluation, kobest_hellaswag does not match, although other datasets(kobest_copa, wic, boolq) match well.
Is there a problem or the kobest_hellaswage data has been changed?

rycont

Nov 28, 2023

•

edited Nov 28, 2023

저는 skt/kogpt-trinity-1.2b-v0.5 모델을 lm-evaluation-harness로 평가해보았으나, 유사하게 HellaSwag와 WiC 태스크에서 상이한 스코어가 나왔습니다. (5-shot / F1 Score 기준으로) HellaSwag와 WiC는 원문에선 각각 0.5272, 0.4313점을 받았다고 하지만, 제가 테스트한 결과로는 0.3999와 0.3953로 큰 차이가 났습니다. 자세한 내용은 다음 블로그 포스트에 적어두었습니다.

https://tilnote.io/pages/65649f20f097f21e37a1c1ad

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment