upload

momozzing · Dec 4, 2023 · ebe3f8b · ebe3f8b
1 parent dcf3c10
commit ebe3f8b
Showing 1 changed file with 2 additions and 3 deletions.
diff --git a/_posts/2023-12-03-효율적인 LLM 학습전략.md b/_posts/2023-12-03-효율적인 LLM 학습전략.md
@@ -52,10 +52,9 @@ Quantization이란?
 
 기존의 언어모델의 파라미터는 Float32로 표현되었는데, 이를 FP16, BF16, FP8, NF4 등 같은 Low bit로 압축을 한다. 
 
-![Alt text](image.png)
-
-![Alt text](image-1.png)
+![image](https://github.com/momozzing/KLUE-TOD/assets/60643542/04672591-7ef7-48d4-94c9-6edfef2a35a1)
 
+![image](https://github.com/momozzing/KLUE-TOD/assets/60643542/105db370-8b01-481e-8551-2ce1cdeb9df3)
 
 ## **Parameter-Efficient Fine-Tuning (PEFT)**