Instructions to use squarelike/llama-2-ko-story-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use squarelike/llama-2-ko-story-7b with Transformers:

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

pipe = pipeline("text-generation", model="squarelike/llama-2-ko-story-7b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("squarelike/llama-2-ko-story-7b")
model = AutoModelForCausalLM.from_pretrained("squarelike/llama-2-ko-story-7b", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use squarelike/llama-2-ko-story-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "squarelike/llama-2-ko-story-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "squarelike/llama-2-ko-story-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/squarelike/llama-2-ko-story-7b

SGLang

How to use squarelike/llama-2-ko-story-7b 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 "squarelike/llama-2-ko-story-7b" \
    --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": "squarelike/llama-2-ko-story-7b",
		"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 "squarelike/llama-2-ko-story-7b" \
        --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": "squarelike/llama-2-ko-story-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use squarelike/llama-2-ko-story-7b with Docker Model Runner:
```
docker model run hf.co/squarelike/llama-2-ko-story-7b
```

llama-2-ko-story-7b

llama-2-ko-story-7b는 beomi/llama-2-ko-7b를 기반으로 한글 소설 raw 데이터를 학습시킨 기반 모델입니다.

학습 데이터

llama-2-ko-story-7b는 약 167MB의 한글 소설 말뭉치로 학습되었습니다. 주요 데이터셋은 다음과 같습니다.

Source	Size (MB)	Link
한글 소설 말뭉치	115.0
공유마당 한국 고전 문학 말뭉치	53.0	https://gongu.copyright.or.kr/

학습

llama-2-ko-story-7b는 beomi/llama-2-ko-7b에서 qlora로 추가 학습되었습니다.

lora_alpha: 16
lora_dropout: 0.05
lora_r: 32
target_modules: q_proj, v_proj
epoch: 3
learning_rate: 3e-4

Downloads last month: 9

Safetensors

Model size

7B params

Tensor type

F32