Update README.md

README.md

@@ -58,7 +58,7 @@ This model predicts strength of causation (SoC) labels based on the text inputs
 
 # Performance and hyperparameters
 ## Test metrics
-This model achieves the following results on the test dataset. The test dataset is a 25% held-out split of the entire dataset with `SEED=114514`.
+This model achieves the following results on the test dataset. The test dataset is a 25% held-out stratified split of the entire dataset with `SEED=114514`.
  * Loss: 0.0049709
  * Off-by-1 accuracy: 100.0000
  * Off-by-2 accuracy: 100.0000