Update README.md
Browse files
README.md
CHANGED
@@ -31,8 +31,8 @@ AI ์ ๋น
๋ฐ์ดํฐ ๋ถ์ ์ ๋ฌธ ๊ธฐ์
์ธ Linkbricks์ ๋ฐ์ดํฐ์ฌ์ด์ธํฐ
|
|
31 |
ํ๊ธ ์ธ์ด ๋ชจ๋ธ๋ก ํ๊ตญ์ด-์ค๊ตญ์ด-์์ด-์ผ๋ณธ์ด ๊ต์ฐจ ํ์ต ๋ฐ์ดํฐ์ ๋ก์ง์ปฌ ๋ฐ์ดํฐ๋ฅผ ํตํ์ฌ ํ์ค์ผ์ ์ธ์ด ๊ต์ฐจ ์ฆ๊ฐ ์ฒ๋ฆฌ์ ๋ณต์กํ ํ๊ธ ๋
ผ๋ฆฌ ๋ฌธ์ ์ญ์ ๋์ ๊ฐ๋ฅํ๋๋ก ํ๋ จํ ๋ชจ๋ธ์ด๋ฉฐ ํ ํฌ๋์ด์ ๋ ๋จ์ด ํ์ฅ ์์ด ๋ฒ ์ด์ค ๋ชจ๋ธ ๊ทธ๋๋ก ์ฌ์ฉ.
|
32 |
ํนํ ๊ณ ๊ฐ ๋ฆฌ๋ทฐ๋ ์์
ํฌ์คํ
๊ณ ์ฐจ์ ๋ถ์ ๋ฐ ์ฝ๋ฉ๋ฑ์ด ๊ฐํ๋ ๋ชจ๋ธ
|
33 |
Deepspeed Stage=3, rslora ๋ฅผ ์ฌ์ฉ <br>
|
34 |
-
ollama run benedict/linkbricks-mixtral-8x7-korean-q4 <br>
|
35 |
-
ollama run benedict/linkbricks-mixtral-8x7-korean-q8
|
36 |
|
37 |
Dr. Yunsung Ji (Saxo), a data scientist at Linkbricks, a company specializing in AI and big data analytics, fine-tuned the NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO base model with SFT->DPO using four H100-80Gs on KT-CLOUD.
|
38 |
It is a Korean language model trained to handle complex Korean logic problems through Korean-Chinese-English-Japanese cross-training data and logical data, and Tokenizer uses the base model without word expansion.
|
|
|
31 |
ํ๊ธ ์ธ์ด ๋ชจ๋ธ๋ก ํ๊ตญ์ด-์ค๊ตญ์ด-์์ด-์ผ๋ณธ์ด ๊ต์ฐจ ํ์ต ๋ฐ์ดํฐ์ ๋ก์ง์ปฌ ๋ฐ์ดํฐ๋ฅผ ํตํ์ฌ ํ์ค์ผ์ ์ธ์ด ๊ต์ฐจ ์ฆ๊ฐ ์ฒ๋ฆฌ์ ๋ณต์กํ ํ๊ธ ๋
ผ๋ฆฌ ๋ฌธ์ ์ญ์ ๋์ ๊ฐ๋ฅํ๋๋ก ํ๋ จํ ๋ชจ๋ธ์ด๋ฉฐ ํ ํฌ๋์ด์ ๋ ๋จ์ด ํ์ฅ ์์ด ๋ฒ ์ด์ค ๋ชจ๋ธ ๊ทธ๋๋ก ์ฌ์ฉ.
|
32 |
ํนํ ๊ณ ๊ฐ ๋ฆฌ๋ทฐ๋ ์์
ํฌ์คํ
๊ณ ์ฐจ์ ๋ถ์ ๋ฐ ์ฝ๋ฉ๋ฑ์ด ๊ฐํ๋ ๋ชจ๋ธ
|
33 |
Deepspeed Stage=3, rslora ๋ฅผ ์ฌ์ฉ <br>
|
34 |
+
ollama run benedict/linkbricks-mixtral-8x7-korean-q4:56b <br>
|
35 |
+
ollama run benedict/linkbricks-mixtral-8x7-korean-q8:56b
|
36 |
|
37 |
Dr. Yunsung Ji (Saxo), a data scientist at Linkbricks, a company specializing in AI and big data analytics, fine-tuned the NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO base model with SFT->DPO using four H100-80Gs on KT-CLOUD.
|
38 |
It is a Korean language model trained to handle complex Korean logic problems through Korean-Chinese-English-Japanese cross-training data and logical data, and Tokenizer uses the base model without word expansion.
|