NAPS-ai/naps-llama-3_1-8b-instruct-v0.4

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

unsloth를 사용하여 meta-llama/Meta-Llama-3.1-8B-Instruct 모델에 LORA 파인튜닝을 완료했습니다.

MarkrAI/KOpen-HQ-Hermes-2.5-60k 데이터를 학습시켰습니다.

How to use

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("NAPS-ai/naps-llama-3.1-8b-instruct-v0.4")
model = AutoModelForCausalLM.from_pretrained("NAPS-ai/naps-llama-3.1-8b-instruct-v0.4")

Chatbot

from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch

tokenizer = AutoTokenizer.from_pretrained("NAPS-ai/naps-llama-3.1-8b-instruct-v0.4")
model = AutoModelForCausalLM.from_pretrained("NAPS-ai/naps-llama-3.1-8b-instruct-v0.4")




pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    model_kwargs={"torch_dtype": torch.bfloat16},
    device=0,
)

def answering(question):
    messages = [
    {"role": "system", "content": "당신은 항상 친절하게 대답하는 안내원입니다."},
    {"role": "user", "content": question},
    ]
    outputs = pipeline(
    messages,
    max_new_tokens=1024,
    pad_token_id = pipeline.tokenizer.eos_token_id
    )
    return outputs[0]["generated_text"][2]['content']




while True:
    question = input("질문을 입력하세요 : ")
    if question == "종료":
        print("프로그램 종료")
        break
    answer = answering(question)
    print(f"AI의 답변: {answer}")

Contact : cgh@tnap.co.kr