Instructions to use cooler8/yejin-korean-1b-v8-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use cooler8/yejin-korean-1b-v8-sft with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf cooler8/yejin-korean-1b-v8-sft:F16 # Run inference directly in the terminal: llama cli -hf cooler8/yejin-korean-1b-v8-sft:F16
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf cooler8/yejin-korean-1b-v8-sft:F16 # Run inference directly in the terminal: llama cli -hf cooler8/yejin-korean-1b-v8-sft:F16
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf cooler8/yejin-korean-1b-v8-sft:F16 # Run inference directly in the terminal: ./build/bin/llama cli -hf cooler8/yejin-korean-1b-v8-sft:F16
Use Docker
docker model run hf.co/cooler8/yejin-korean-1b-v8-sft:F16
- Llama new
- LM Studio
- Jan
- vLLM
How to use cooler8/yejin-korean-1b-v8-sft with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "cooler8/yejin-korean-1b-v8-sft" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cooler8/yejin-korean-1b-v8-sft", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/cooler8/yejin-korean-1b-v8-sft:F16
- Ollama
How to use cooler8/yejin-korean-1b-v8-sft with Ollama:
ollama run hf.co/cooler8/yejin-korean-1b-v8-sft:F16
- Unsloth Desktop
- Docker Model Runner
How to use cooler8/yejin-korean-1b-v8-sft with Docker Model Runner:
docker model run hf.co/cooler8/yejin-korean-1b-v8-sft:F16
- Lemonade
How to use cooler8/yejin-korean-1b-v8-sft with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull cooler8/yejin-korean-1b-v8-sft:F16
Run and chat with the model
lemonade run user.yejin-korean-1b-v8-sft-F16
List all available models
lemonade list
- Atomic Chat
yejin-korean-1b-v8-sft π
yejin-korean-1b-v8-sftλ νκ΅μ΄ 172GB+ κ³ νμ§ μ½νΌμ€λ‘ 100% λ°λ₯λΆν° μ¬μ νμ΅(Pre-trained from Scratch)λ
yejin-korean-1b-v8λ² μ΄μ€ λͺ¨λΈμ KMMLU, KoAlpaca, νκ΅μ΄ κ΅κ³Όμ μ§μλ¬Έμ κ²°ν©νμ¬ μ§λ λ―ΈμΈμ‘°μ (Supervised Fine-Tuning)μ μλ£ν 1B νκ΅μ΄ μν νμ΄λ°μ΄μ λͺ¨λΈμ λλ€.
π λͺ¨λΈ μν€ν μ² μ¬μ
- κΈ°λ° κ·κ²©: Llama 3 μ°Έμ‘° μν€ν
μ² (
hidden=2048,layers=18,heads=16,kv_heads=4,intermediate=5632,GQA 4:1,RoPE ΞΈ=10000) - νλΌλ―Έν° μ: μ½ 9μ΅ 3,200λ§ (0.93B)
- ν ν¬λμ΄μ :
EleutherAI/polyglot-ko-1.3b(μ΄ν ν¬κΈ°: 30,003) - 컨ν μ€νΈ κΈΈμ΄: 4,096 ν ν°
- ν¬λ§·:
safetensors(Float32 / BFloat16)
π μ¬μ© μμ (Python Transformers)
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
model_id = "cooler8/yejin-korean-1b-v8-sft"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float32, device_map="auto")
prompt = "### μ§μμ¬ν:\nμΈκ³΅μ§λ₯μ΄λ 무μμΈμ§ μ΄λ±νμλ μκΈ° μ½κ² μ€λͺ
ν΄μ€.\n\n### λ΅λ³:\n"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output = model.generate(
input_ids=inputs["input_ids"],
attention_mask=inputs["attention_mask"],
max_new_tokens=200,
temperature=0.7,
top_p=0.9,
repetition_penalty=1.15
)
print(tokenizer.decode(output[0], skip_special_tokens=True))
π‘ ν둬ννΈ ν¬λ§· (Prompt Template)
### μ§μμ¬ν:
{μ¬μ©μ μ§λ¬Έ λλ μ§μλ¬Έ}
### λ΅λ³:
{λͺ¨λΈ μμ± λ΅λ³}
- Downloads last month
- 36