Update README.md

README.md

@@ -22,7 +22,7 @@ model-index:
 
 # vit-base-nsfw-detector
 
-This model is a fine-tuned version of [vit-base-patch16-384](https://huggingface.co/google/vit-base-patch16-384) on around 2000 images (drawings, photos...).
+This model is a fine-tuned version of [vit-base-patch16-384](https://huggingface.co/google/vit-base-patch16-384) on around 25_000 images (drawings, photos...).
 It achieves the following results on the evaluation set:
 - Loss: 0.0937
 - Accuracy: 0.9654
@@ -89,7 +89,7 @@ The following hyperparameters were used during training:
 - Accuracy: 0.9654, 
 - AUC: 0.9948
 
-Confusion Matrix:
+Confusion Matrix (eval):
 
 [1076   37]