edited typo

wonchul-kim · Jan 6, 2024 · e27e39d · e27e39d
1 parent a6c96e2
commit e27e39d
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/_posts/papers/language_model/attention_is_all_you_need/2024-01-06-papers.md b/_posts/papers/language_model/attention_is_all_you_need/2024-01-06-papers.md
@@ -14,7 +14,7 @@ tag: language  attention
 <img src='./assets/attention_2.gif'>
 <img src='{{ "/_posts/papers/languge_model/attention_is_all_you_need/assets/attention_2.gif" | prepend: site.baseurl }}'>
 
-![image-title-here](/assets/attention_2.gif){:class="img-responsive"}
+![image-title-here](_posts/papers/languge_model/attention_is_all_you_need/assets/attention_2.gif){:class="img-responsive"}
 
 1. `RNN`으로 인해서 지속적으로 `encoder`의 `state`에 대한 모든 처리가 순차적으로 필요한 `seq2seq` 개념으로 계산 속도가 느림.