File size: 1,248 Bytes
60f0e74
66dad03
 
 
 
 
 
 
 
 
 
60f0e74
 
66dad03
60f0e74
66dad03
60f0e74
66dad03
 
 
 
 
60f0e74
66dad03
60f0e74
66dad03
 
 
60f0e74
66dad03
 
 
 
 
 
60f0e74
66dad03
 
 
 
 
 
 
60f0e74
66dad03
 
 
60f0e74
66dad03
 
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
---
license: apache-2.0
base_model: kakaocorp/kanana-safeguard-prompt-2.1b
quantized_by: Arc1el
quantization_method: bitsandbytes
tags:
- quantized
- 4bit
- bitsandbytes
- safeguard
- korean
---

# Kanana Safeguard Prompt 2.1B - 4bit Quantized

4bit ์–‘์žํ™”๋œ kakaocorp/kanana-safeguard-prompt-2.1b ๋ชจ๋ธ์ž…๋‹ˆ๋‹ค.

## ๋ชจ๋ธ ์ •๋ณด
- **์›๋ณธ ๋ชจ๋ธ**: [kakaocorp/kanana-safeguard-prompt-2.1b](https://huggingface.co/kakaocorp/kanana-safeguard-prompt-2.1b)
- **์–‘์žํ™” ๋ฐฉ๋ฒ•**: BitsAndBytes 4bit (NF4)
- **ํŒŒ๋ผ๋ฏธํ„ฐ**: 2.1B
- **์šฉ๋„**: ํ”„๋กฌํ”„ํŠธ ์•ˆ์ „์„ฑ ๊ฒ€์ฆ

## ์‚ฌ์šฉ๋ฒ•

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_use_double_quant=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16
)

model = AutoModelForCausalLM.from_pretrained(
    "[your-username]/kanana-safeguard-prompt-2.1b-4bit",
    quantization_config=bnb_config,
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("nxtcloud-org/kanana-safeguard-prompt-2.1b-4bit")
```

## ์‹œ์Šคํ…œ ์š”๊ตฌ์‚ฌํ•ญ
- GPU ๋ฉ”๋ชจ๋ฆฌ: 2GB+
- RAM: 4GB+

## ๋ผ์ด์„ ์Šค
Apache License 2.0 (์›๋ณธ ๋ชจ๋ธ๊ณผ ๋™์ผ)