Model Details

기존 meta-llama/Meta-Llama-3.1-8B-Instruct 모델의 32개 layer중 10개 layer를 삭제하고 학습한 모델입니다

Uses

import transformers
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kikikara/ko-llama-3.1-5b-instruct")
model = AutoModelForCausalLM.from_pretrained("kikikara/ko-llama-3.1-5b-instruct", device_map="auto")

pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    device_map="auto",
)

question = "왜 살아야 하는지 철학적 측면에서 접근해봐"
messages = [
    {"role": "system", "content": "당신은 한국어 ai 모델입니다."},
    {"role": "user", "content": question},
]

outputs = pipeline(
    messages,
    repetition_penalty=1.1,
    max_new_tokens=1500,
)

print(outputs[0]["generated_text"][-1]['content'])
Downloads last month
2
Safetensors
Model size
6B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for kikikara/ko-llama-3.1-5b-instruct

Quantizations
1 model