Edit model card

https://github.com/jwj7140/ko-medical-chat

Llama-Ko-Medical-7b

llama2-ko-medical์€ llama-2-ko๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ์˜๋ฃŒ ๋ถ„์•ผ์˜ ํ•œ๊ธ€ raw ๋ฐ์ดํ„ฐ๋ฅผ ํ•™์Šต์‹œํ‚จ ๊ธฐ๋ฐ˜ ๋ชจ๋ธ์ž…๋‹ˆ๋‹ค.

ํ•™์Šต ๋ฐ์ดํ„ฐ

llama2-ko-medical์€ ์•ฝ 526MB์˜ ์˜๋ฃŒ ๋ถ„์•ผ ํ•œ๊ธ€ ๋ง๋ญ‰์น˜๋กœ ํ•™์Šต๋˜์—ˆ์Šต๋‹ˆ๋‹ค. ์ฃผ์š” ๋ฐ์ดํ„ฐ์…‹์€ ๋‹ค์Œ๊ณผ ๊ฐ™์Šต๋‹ˆ๋‹ค.

Source Size (MB) Link
AIHub ์˜๋ฃŒ, ๋ฒ•๋ฅ  ์ „๋ฌธ ์„œ์  ๋ง๋ญ‰์น˜ 351.0 aihub.or.kr
ํ•˜์ด๋‹ฅ ๋‰ด์Šค ๋ฐ์ดํ„ฐ 97.3 hidoc.co.kr
AIHub ์ „๋ฌธ๋ถ„์•ผ ํ•œ์˜ ๋ง๋ญ‰์น˜ 63.4 aihub.or.kr
์งˆ๋ณ‘๊ด€๋ฆฌ์ฒญ ๊ตญ๊ฐ€๊ฑด๊ฐ•์ •๋ณดํฌํ„ธ 8.33 health.kdca.go.kr
๋ณด๊ฑด๋ณต์ง€๋ถ€ ๊ตญ๊ฐ€์ •์‹ ๊ฑด๊ฐ•์ •๋ณดํฌํ„ธ < 1.0 mentalhealth.go.kr

ํ•™์Šต

llama2-ko-medical-7b๋Š” llama-2-ko์—์„œ qlora๋กœ ์ถ”๊ฐ€ ํ•™์Šต๋˜์—ˆ์Šต๋‹ˆ๋‹ค.

  • lora_alpha: 16
  • lora_dropout: 0.01
  • lora_r: 64
  • target_modules: q_proj, v_proj
  • epoch: 3
  • learning_rate: 3e-4
Downloads last month
4,486
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for squarelike/llama2-ko-medical-7b

Adapters
2 models