File size: 9,428 Bytes
d9eb4eb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 |
---
license: gemma
tags:
- bccard
- bcgpt
datasets:
- BCCard/BCCard-Finance-Kor-QnA
language:
- ko
- en
- ja
base_model:
- google/gemma-2-27b-it
---
**BCCard/gemma-2-27b-it-kor-BCGPT** is a Korean and Financial knowledge model based on Google Gemma 2 27B.
BC Card, which is the largest credit card company in Korea, is a question/answer model learned using Korean financial datasets.
# How to Use
```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
model_id = "BCCard/gemma-2-27b-it-kor-BCGPT"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
model_id,
torch_dtype=torch.bfloat16,
device_map="auto",
)
model.eval()
instruction = "์์ธ์ ์ ๋ช
ํ ๊ด๊ด ์ฝ์ค๋ฅผ ๋ง๋ค์ด์ค๋?"
messages = [
{"role": "user", "content": f"{instruction}"}
]
input_ids = tokenizer.apply_chat_template(
messages,
add_generation_prompt=True,
return_tensors="pt"
).to(model.device)
terminators = [
tokenizer.eos_token_id,
tokenizer.convert_tokens_to_ids("<end_of_turn>")
]
outputs = model.generate(
input_ids,
max_new_tokens=2048,
eos_token_id=terminators,
do_sample=True,
temperature=0.6,
top_p=0.9,
)
print(tokenizer.decode(outputs[0][input_ids.shape[-1]:], skip_special_tokens=True))
```
# Generated Text
```
์์ธ์ ์ ๋ช
ํ ๊ด๊ด ์ฝ์ค๋ฅผ ๋ง๋ค์ด๋ณผ๊ฒ์.
1. **์ญ์ฌ์ ๋ฌธํ๋ฅผ ํ๋ฐฉํ๋ ์ฝ์ค:** ๊ฒฝ๋ณต๊ถ, ์ฐฝ๋๊ถ, ๊ตญ๋ฆฝ๋ฏผ์๋ฐ๋ฌผ๊ด, ์ธ์ฌ๋, ์ข
๋ก3๊ฐ, ๋จ์ฐ์ ํฌํจํ ์ด ์ฝ์ค๋ ์์ธ์ ์ญ์ฌ์ ๋ฌธํ๋ฅผ ๊น์ด ํ๊ตฌํ ์ ์๋ ๋ค์ํ ์ฅ์๋ฅผ ๋ฐฉ๋ฌธํ ์ ์๋๋ก ๊ตฌ์ฑ๋ฉ๋๋ค. ์ญ์ฌ์ ๋ฌธํ๋ฅผ ํ๋ฐฉํ๋ ์ด ์ฝ์ค๋ ์์ธ์ ์ญ์ฌ๋ฅผ ์ดํดํ๊ณ , ํ๊ตญ์ ์ ํต๊ณผ ํ๋๋ฅผ ๋์์ ์ฆ๊ธธ ์ ์์ต๋๋ค.
2. **K-ํ๊ณผ ์ผํ์ ์ฆ๊ธฐ๋ ์ฝ์ค:** ์ด ์ฝ์ค๋ ํ๊ตญ์์ ๊ฐ์ฅ ์ธ๊ธฐ ์๋ K-ํ ์์ ์ธ 'K-Star Road'๋ฅผ ๋ฐฉ๋ฌธํ๋ ์ฝ์ค๋ก, ์ด๊ณณ์์ ์ผํ์ ํ๊ณ ๋จน๊ฑฐ๋ฆฌ๋ฅผ ์ฆ๊ธฐ๋ K-ํ ํฌ๋ค์ด ๋ง์ด ์ฐพ๋ ๊ณณ์
๋๋ค. ์ด ์ฝ์ค๋ ํฌ๋ค์ด ํ๊ตญ์ ํซํ๋ ์ด์ค์์ ์ฆ๊ธธ ์ ์๋ ์ผํ๊ณผ ํ๊ตญ์ K-ํ ๋ฌธํ๋ฅผ ์ง์ ์ฒดํํ ์ ์๋ ๊ธฐํ๋ฅผ ์ ๊ณตํฉ๋๋ค.
3. **ํธ๋ ๋ํ ์นดํ์ ์์์ ์ ๋๋ฌ๋ณด๋ K-ํจ์ค ์นด๋:** ์์ธ์์ ๊ฐ์ฅ ์ธ๊ธฐ ์๋ ์นดํ๋ ๋ ์คํ ๋์ ์ฝ๊ฒ ์ด๋ํ ์ ์๋ ๋์ค๊ตํต ํจ์ค๋ฅผ ์ ๊ณตํ๋ ์นด๋๋ ์์ต๋๋ค. ์ด๋ ๋์ค๊ตํต์ ์ด์ฉํ๋ ์ฌ์ฉ์์๊ฒ ๋งค์ฐ ์ ๋ฆฌํฉ๋๋ค.
```
-------
# Q&A
```
Q. ํ์ด๋ถ ๋จธ๋์ ๋ํ ์ค๋ช
ํด์ฃผ์ธ์.
A. ํ์ด๋ถ ๋จธ๋๋ BC์นด๋์ ๊ฐํธ๊ฒฐ์ ์๋น์ค ํ์ด๋ถ์์ ์ฌ์ฉ๋๋ ํฌ์ธํธ ์ ๋ฆฝ ๋ฐ ํ ์ธ ํํ์ ์ ๊ณตํ๋ BC์นด๋์ ํฌ์ธํธ ์ ๋ก ์๋น์ค๋ฅผ ํตํด ์ ๋ฆฝ๋ ํฌ์ธํธ์
๋๋ค. ํ์ด๋ถ ๋จธ๋๋ BC์นด๋์ ๋จธ๋ ์๋น์ค๋ก, ์ผํ, ์ํ ํํ์ ์ ๊ณตํ๋๋ฐ ์ฌ์ฉ๋๋ ๋์งํธ ํฌ์ธํธ ๋ฐ ์ถฉ์ ์ ์ ์ ์ง๊ธ์๋จ์ ์๋ฏธํฉ๋๋ค.
Q. ์นด๋ ์ฐ์ฒดํ๋ฉด ์ด๋ป๊ฒ ๋ผ? ๋ ์ ๊ฐ๊ณ ์ถ์ด. ์ ๊ฐ์์ผ ํ๋์ง ์์ธํ ๋งํด๋ด.
A. ์ ์ฉ์นด๋ ์ฐ์ฒด๋ ์ ์ฉ ์ ์์ ๋ถ์ ์ ์ธ ์ํฅ์ ๋ฏธ์น๋ฉฐ, ์ด๋ ๊ธ์ต ์ํ์ ํฐ ๋ถํธ์ ์ด๋ํ ์ ์์ต๋๋ค. ์ฐ์ฒด๊ฐ ๋ฐ์ํ๋ฉด ์ ์ฉ์นด๋ ํ์ฌ๋ ์ฐ์ฒด ์ด์์ ์ฐ์ฒด๋ฃ๋ฅผ ๋ถ๊ณผํ ์ ์์ผ๋ฉฐ, ์ด๋ ์ฐ์ฒด ๊ธฐ๊ฐ ๋์์ ํ๊ท ์์ก์ ๋ฐ๋ผ ๋ฌ๋ผ์ง๋๋ค.
Q. ์นด๋ ํ ๋ถ๋ฅผ ํ๋๋ฐ ๊ฒฐ์ ํ ๊ฐ๋งน์ ์ด ๋งํ์ด. ๋ณด์ ๋ฐ์ ์ ์์ด?
A. ๊ฐ๋งน์ ์ด ๋งํ์ ๊ฒฝ์ฐ ์นด๋์ฌ์ ๋ฌธ์ํ์ฌ ํ ๋ถ๊ธ์ ์ทจ์ํ ์ ์์ต๋๋ค. ์นด๋์ฌ๋ ์๋น์๊ฐ ํผํด๋ฅผ ๋ณด์ง ์๋๋ก ์ผ์ ์กฐ๊ฑด ํ์ ํ ๋ถ ์ทจ์๋ฅผ ํ์ฉํ๋ฉฐ, ์ด๋ฏธ ๋ฉ๋ถํ ๊ธ์ก์ ๋ํด์๋ ํ๋ถ ์ ์ฐจ๋ฅผ ์งํํ ์ ์์ต๋๋ค. ๋จ, ํ๋ถ ์ ์ฐจ๋ ๊ฐ๋งน์ ์ ์ ์ฐ ์ํ ๋ฑ์ ๋ฐ๋ผ ๋ค์ ์๊ฐ์ด ๊ฑธ๋ฆด ์ ์์ต๋๋ค.
Q. K-ํจ์ค ์นด๋ ํ ์ธ ํํ๋ ์๋ ค์ค๋ด
A. ๋น์จ K-ํจ์ค ์นด๋๋ ๋์ค๊ตํต ์ด์ฉ ์ 15% ํ ์ธ์ ์ ๊ณตํ๋ฉฐ, ์ฃผ์ OTT ์๋น์ค์ ์ํ๊ด ํ ์ธ ํํ๋ ํฌํจ๋์ด ์์ต๋๋ค. ๋ํ, K-์ปฌ์ฒ ๊ด๋ จ ์
์ข
์์๋ ๋ค์ํ ํ ์ธ ํํ์ ๋ฐ์ ์ ์์ต๋๋ค.
Q. ๊ทธ๋ฆฐ์นด๋์ ๋ํด์ ์์ธํ ์ค๋ช
ํด์ฃผ์ธ์.
A. ## ๊ทธ๋ฆฐ์นด๋: ์นํ๊ฒฝ ์๋น๋ก ํํ์ ๋ฐ๋ ์นด๋
๊ทธ๋ฆฐ์นด๋๋ ์๋์ง ์ ์ฝ ๋ฐ ํ๊ฒฝ ๋ณดํธ ํ๋์ ์ฐธ์ฌํ ๋ ํํ์ ๋ฐ์ ์ ์๋ ์นด๋๋ก, ๋น์จ์นด๋, ์ ํ์นด๋, ์ผ์ฑ์นด๋, KB๊ตญ๋ฏผ์นด๋, ์ฐ๋ฆฌ์นด๋, ํ๋์นด๋ ๋ฑ ์ฌ๋ฌ ์นด๋์ฌ์์ ๋ฐ๊ธํ๊ณ ์์ต๋๋ค. ์ด ์นด๋๋ฅผ ํตํด ์นํ๊ฒฝ ์๋น๋ฅผ ์ค์ฒํ๋ฉด์ ๋ค์ํ ํํ์ ๋๋ฆด ์ ์์ต๋๋ค.
### ์ฃผ์ ํํ
1. **ํ์ํฌ์ธํธ ์ ๋ฆฝ**: ์นํ๊ฒฝ ๋งค์ฅ์์ ๊ฒฐ์ ํ๊ฑฐ๋ ๋์ค๊ตํต ์ด์ฉ ์ ํ์ํฌ์ธํธ๋ฅผ ์ ๋ฆฝํ ์ ์์ต๋๋ค. ์๋ฅผ ๋ค์ด, ์ ๊ธฐ์ฐจ ์ถฉ์ ์์์ ๊ฒฐ์ ์ ํฌ์ธํธ ์ ๋ฆฝ๋ฅ ์ด ๋์ต๋๋ค.
2. **๊ทธ๋ฆฐ์๋น์ค ์ด์ฉ**: ๊ทธ๋ฆฐ์นด๋๋ฅผ ํตํด ๋ค์ํ ๊ทธ๋ฆฐ์๋น์ค๋ฅผ ํ ์ธ๋ ๊ฐ๊ฒฉ์ ์ด์ฉํ ์ ์์ต๋๋ค. ์๋ฅผ ๋ค์ด, ์ ๊ธฐ์ฐจ ์ถฉ์ ์ ํ ์ธ, ๊ณต์ ๋ชจ๋น๋ฆฌํฐ ์ด์ฉ ์ ํ ์ธ ๋ฑ์ด ์์ต๋๋ค.
3. **ํฌ์ธํธ ๊ตํ**: ์ ๋ฆฝ๋ ํ์ํฌ์ธํธ๋ ํ๊ธ์ฒ๋ผ ์ฌ์ฉํ ์ ์์ผ๋ฉฐ, ํน์ ๊ธฐ๋ถ์ฒ์ ๊ธฐ๋ถํ ์๋ ์์ต๋๋ค.
### ์์ฝ๋จธ๋ ํฌ์ธํธ ์ ๋ฆฝ ๋ฐฉ๋ฒ
1. **์จ๋ผ์ธ ์ผํ**: ์์ฝ๋จธ๋ ์ ํด ์ผํ๋ชฐ์์ ์นํ๊ฒฝ ์ธ์ฆ ์ ํ์ ๊ตฌ๋งคํ๋ฉด 1%~10%์ ์์ฝ๋จธ๋ ํฌ์ธํธ๋ฅผ ์ ๋ฆฝํ ์ ์์ต๋๋ค.
2. **์คํ๋ผ์ธ ๋งค์ฅ**: ์์ฝ๋จธ๋ ์ ํด ๋งค์ฅ์์ ์นํ๊ฒฝ ์ ํ์ ๊ตฌ๋งคํ๊ฑฐ๋ ์์ฝ๋จธ๋ ๊ฐ๋งน์ ์์ ๊ฒฐ์ ํ๋ฉด ํฌ์ธํธ๊ฐ ์ ๋ฆฝ๋ฉ๋๋ค.
3. **๋์ค๊ตํต ์ด์ฉ**: ๋ฒ์ค, ์งํ์ฒ , ํ์ ๋ฑ ๋์ค๊ตํต์ ์ด์ฉํ๋ฉด ์ฌ์ฉ ๊ธ์ก์ ์ผ์ ๋น์จ์ ํฌ์ธํธ๋ก ์ ๋ฆฝํ ์ ์์ต๋๋ค.
4. **์๋์ง ์ ์ฝ**: ๊ฐ์ ์ ํ ์ฌ์ฉ๋์ ์ค์ด๊ฑฐ๋ LED ์ ๊ตฌ๋ฅผ ์ค์นํ์ฌ ์๋์ง๋ฅผ ์ ์ฝํ๋ฉด ํฌ์ธํธ๋ฅผ ์ ๋ฆฝํ ์ ์์ต๋๋ค.
### ์ด์
1. **ํ๊ฒฝ ๋ณดํธ**: ์นํ๊ฒฝ ์๋น๋ฅผ ์ฅ๋ คํ์ฌ ํ๊ฒฝ ๋ณดํธ์ ๊ธฐ์ฌํ ์ ์์ต๋๋ค.
2. **๊ฒฝ์ ์ ํํ**: ๋ค์ํ ํ ์ธ ํํ๊ณผ ํฌ์ธํธ ์ ๋ฆฝ์ ํตํด ๊ฒฝ์ ์ ์ด์ต์ ์ป์ ์ ์์ต๋๋ค.
3. **์ฌํ์ ์ฑ
์**: ์นํ๊ฒฝ ์ ํ์ ์ฌ์ฉํ๊ณ ์๋์ง๋ฅผ ์ ์ฝํ๋ ๋ฑ ์ฌํ์ ์ฑ
์์ ๋คํ ์ ์์ต๋๋ค.
### ์ฌ์ฉ ๋ฐฉ๋ฒ
๊ทธ๋ฆฐ์นด๋๋ฅผ ์ฌ์ฉํ๋ฉด, ์นํ๊ฒฝ ์ธ์ฆ ์ ํ์ ๊ตฌ๋งคํ๊ฑฐ๋ ์๋์ง๋ฅผ ์ ์ฝํ๋ ๋ฑ์ ํ๋์ ํตํด ํฌ์ธํธ๋ฅผ ์ ๋ฆฝํ ์ ์์ต๋๋ค. ์ ๋ฆฝ๋ ํฌ์ธํธ๋ ํ๊ธ์ฒ๋ผ ์ฌ์ฉํ๊ฑฐ๋ ๊ธฐ๋ถํ ์ ์์ต๋๋ค.
Q. ์ผ์ด๋ฑ
ํฌ์์ ๋์ถ์ ๋ฐ๊ณ ์ถ์๋ฐ, ์นด๋๋ก 300๋ง์์ ๋ฐ์์ด, ๋์ถ์ ๋ฏธ์น๋ ์ํฅ์ ๋ํด์ ์๋ ค์ค
A. ์ผ์ด๋ฑ
ํฌ์์ ์นด๋๋ก 300๋ง์์ ์ถ๊ฐ๋ก ๋ฐ์์ ๊ฒฝ์ฐ, ๋์ถ ํ๋์ ์ํฅ์ ๋ฏธ์น ์ ์์ต๋๋ค. ๊ฐ ๋์ถ ์ํ์ ๋ณ๋๋ก ๊ด๋ฆฌ๋๋ฏ๋ก, ์ ์ฉ๋์ถ ํ๋์ ์นด๋๋ก ํ๋๋ ๊ฐ๊ฐ ์ ์ฉ๋ฉ๋๋ค. ๋ค๋ง, ์ ์ฒด ๋ถ์ฑ ๊ท๋ชจ๊ฐ ์ปค์ง ๊ฒฝ์ฐ ์ ์ฉ๋ฑ๊ธ์ ์ํฅ์ ์ค ์ ์์ผ๋ฏ๋ก, ์ํ ๋ฅ๋ ฅ์ ๊ณ ๋ คํ์ฌ ์ ์คํ ๊ด๋ฆฌํ๋ ๊ฒ์ด ์ค์ํฉ๋๋ค.
Q. ๋ค์ ๋ด์ฉ์ ์์ด๋ก ๋ฒ์ญํด์ค
์นด๋ ์ฐ์ฒด๋ ์ ์ฉ ์ ์์ ์ง์ ์ ์ธ ์ํฅ์ ๋ฏธ์น๋ฉฐ, ์ฅ๊ธฐ์ ์ผ๋ก ๊ธ์ต ๊ฑฐ๋์ ํฐ ์ ์ฝ์ ์ด๋ํ ์ ์์ต๋๋ค. ์ฐ์ฒด๊ฐ ๋ฐ์ํ๋ฉด ์ฐ์ฒด๋ฃ๊ฐ ๋ถ๊ณผ๋๊ณ , ์ผ๋ถ ์นด๋์ฌ๋ ์ฐ์ฒด ์ ์ถ๊ฐ๋ก ๋์ ์ด์์จ์ ์ ์ฉํฉ๋๋ค. ์๋ฅผ ๋ค์ด, ์ฐ์ฒด ์ด์์จ์ด 20%์ธ ๊ฒฝ์ฐ, ์ฐ์ฒด ๊ธ์ก์ ๋ํด ๋งค๋
20%์ ์ถ๊ฐ ๋น์ฉ์ด ๋ฐ์ํฉ๋๋ค. ๋ํ, ์ฐ์ฒด ์ ๋ณด๊ฐ ์ ์ฉ ๋ณด๊ณ ์์ ๊ธฐ๋ก๋์ด ๋์ถ ์ ์ฒญ์ด๋ ์ถ๊ฐ ์ ์ฉ์นด๋ ๋ฐ๊ธ ์ ๋ถ์ด์ต์ ๋ฐ์ ์ ์์ต๋๋ค. ์ ์ฉ ์ ์๊ฐ ๋ฎ์์ง๋ฉด ๋ณดํ๋ฃ๋ ํต์ ์๊ธ ๋ฑ์์๋ ๋ถ์ด์ต์ด ์์ ์ ์์ต๋๋ค. ๋ฐ๋ผ์, ์นด๋ ๋๊ธ์ ๊ฐ์ง ์๊ธฐ๋ณด๋ค๋ ์ํ ๊ณํ์ ์ฌ์กฐ์ ํ๊ฑฐ๋, ์๋ฏผ๊ธ์ต์งํฅ์์ ์ ๊ธ๋ฆฌ ๋์ถ ์ํ ๋ฑ์ ์ด์ฉํ์ฌ ์ฌ์ ๋ถ๋ด์ ์ค์ด๋ ๊ฒ์ด ๋ฐ๋์งํฉ๋๋ค.
A. Card delinquency has a direct impact on credit scores and can significantly restrict financial transactions in the long run. When delinquency occurs, late fees are imposed, and some card issuers apply higher interest rates to delinquent amounts. For example, if the delinquency interest rate is 20%, an additional annual cost of 20% will be incurred on the delinquent amount. Moreover, delinquency information is recorded in credit reports, potentially leading to disadvantages when applying for loans or additional credit cards. A low credit score can also result in higher insurance premiums and communication costs. Therefore, instead of defaulting on card payments, it is advisable to renegotiate repayment plans or explore financial assistance options like low-interest loans from the Credit Union.
```
## Correspondence to
- Taeyoung Lee ([email protected])
## License
The use of this model is governed by the [GOOGLE GEMMA TERM OF USE](https://ai.google.dev/gemma/terms)
## Citation
If you use this model in your research, please cite it as follows:
```bibtex
@misc{alpha-instruct,
author = {BCCard},
title = {BCGPT},
year = {2025},
publisher = {Hugging Face},
journal = {Hugging Face repository},
url = {https://huggingface.co/BCCard/gemma-2-9b-it-kor-BCGPT},
} |