Instructions to use CarrotAI/Llama3-Ko-Carrot-8B-it with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CarrotAI/Llama3-Ko-Carrot-8B-it with Transformers:

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

pipe = pipeline("text-generation", model="CarrotAI/Llama3-Ko-Carrot-8B-it")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("CarrotAI/Llama3-Ko-Carrot-8B-it")
model = AutoModelForCausalLM.from_pretrained("CarrotAI/Llama3-Ko-Carrot-8B-it")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use CarrotAI/Llama3-Ko-Carrot-8B-it with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CarrotAI/Llama3-Ko-Carrot-8B-it"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CarrotAI/Llama3-Ko-Carrot-8B-it",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CarrotAI/Llama3-Ko-Carrot-8B-it

SGLang

How to use CarrotAI/Llama3-Ko-Carrot-8B-it 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 "CarrotAI/Llama3-Ko-Carrot-8B-it" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CarrotAI/Llama3-Ko-Carrot-8B-it",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "CarrotAI/Llama3-Ko-Carrot-8B-it" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CarrotAI/Llama3-Ko-Carrot-8B-it",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use CarrotAI/Llama3-Ko-Carrot-8B-it with Docker Model Runner:
```
docker model run hf.co/CarrotAI/Llama3-Ko-Carrot-8B-it
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Details

axolotl를 이용하여 공개/자체적으로 생성된 한국어, 영어 데이터셋으로 파인튜닝하였습니다.

Model Description

Socre

llm_kr_eval

평가 지표	점수
AVG_llm_ok_eval	0.4282
EL (Easy Language)	0.1264
FA (False Alarm)	0.2184
NLI (Natural Language Understanding)	0.5767
QA (Question Answering)	0.5100
RC (Reconstruction)	0.7096
klue_ner_set_f1 (Klue Named Entity Recognition F1 Score)	0.1429
klue_re_exact_match (Klue Reference Exact Match)	0.1100
kmmlu_preview_exact_match (Kmmlu Preview Exact Match)	0.4400
kobest_copa_exact_match (Kobest COPA Exact Match)	0.8100
kobest_hs_exact_match (Kobest HS Exact Match)	0.3800
kobest_sn_exact_match (Kobest SN Exact Match)	0.9000
kobest_wic_exact_match (Kobest WIC Exact Match)	0.5800
korea_cg_bleu (Korean CG BLEU)	0.2184
kornli_exact_match (KornLI Exact Match)	0.5400
korsts_pearson (KorSTS Pearson Correlation Coefficient)	0.6225
korsts_spearman (KorSTS Spearman Rank Correlation Coefficient)	0.6064

LogicKor

카테고리	싱글 점수 평균	멀티 점수 평균
수학(Math)	4.43	3.71
이해(Understanding)	9.29	6.86
추론(Reasoning)	5.71	5.00
글쓰기(Writing)	7.86	7.43
코딩(Coding)	7.86	6.86
문법(Grammar)	6.86	3.86
전체 싱글 점수 평균	7.00	-
전체 멀티 점수 평균	-	5.62
전체 점수	-	6.31

Built with Meta Llama 3

License Llama3 License: https://llama.meta.com/llama3/license

Applications

This fine-tuned model is particularly suited for [mention applications, e.g., chatbots, question-answering systems, etc.]. Its enhanced capabilities ensure more accurate and contextually appropriate responses in these domains.

Limitations and Considerations

While our fine-tuning process has optimized the model for specific tasks, it's important to acknowledge potential limitations. The model's performance can still vary based on the complexity of the task and the specificities of the input data. Users are encouraged to evaluate the model thoroughly in their specific context to ensure it meets their requirements.

If you liked this model, please use the card below

@article{Llama3KoCarrot8Bit,
  title={CarrotAI/Llama3-Ko-Carrot-8B-it Card},
  author={CarrotAI (L, GEUN)},
  year={2024},
  url = {https://huggingface.co/CarrotAI/Llama3-Ko-Carrot-8B-it/}
}

Downloads last month: 3

Model tree for CarrotAI/Llama3-Ko-Carrot-8B-it

Quantizations

1 model