Edit model card

Model Details

๊ธฐ์กด meta-llama/Meta-Llama-3.1-8B-Instruct ๋ชจ๋ธ์˜ 32๊ฐœ layer์ค‘ 10๊ฐœ layer๋ฅผ ์‚ญ์ œํ•˜๊ณ  ํ•™์Šตํ•œ ๋ชจ๋ธ์ž…๋‹ˆ๋‹ค

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 = "์™œ ์‚ด์•„์•ผ ํ•˜๋Š”์ง€ ์ฒ ํ•™์  ์ธก๋ฉด์—์„œ ์ ‘๊ทผํ•ด๋ด"
messages = [
    {"role": "system", "content": "๋‹น์‹ ์€ ํ•œ๊ตญ์–ด ai ๋ชจ๋ธ์ž…๋‹ˆ๋‹ค."},
    {"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
22
Safetensors
Model size
5.85B params
Tensor type
F32
ยท
Inference Examples
Inference API (serverless) is not available, repository is disabled.

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

Quantizations
1 model