Update README.md

README.md

@@ -44,9 +44,6 @@ We sampled 16B tokens from the following datasets for training:
   </tr>
 </table>
 
-We trained this model using a context length of 4k due to resource limitations and to maximize training speed.
-However, the original model was trained with a context length of 8k, so an 8k context length could work well in downstream tasks.
-
 ### Hyperparameters
 
 <table>
@@ -142,6 +139,10 @@ We evaluated this model using both English and Korean benchmarks, and compared i
   </tr>
 </table>
 
+## Limitations
+
+We trained this model using a context length of 4k due to resource limitations and to maximize training speed.
+However, the original model was trained with a context length of 8k, so an 8k context length could work well in downstream tasks.
 
 ## License