fix contributions

Author: Dmitrii Tarasov

project/index.html

@@ -165,7 +165,7 @@ <h4>🔍 Novel Feature Analysis Method</h4>
             <div class="box">
               <div class="content">
                 <h4>📊 Model Family Comparison</h4>
-                <p>We reveal that encoders pre-trained on image-based tasks retain significantly more image information compared to those trained on contrastive learning tasks, demonstrated through our SigLIP vs SigLIP2 analysis.</p>
+                <p>We reveal that encoders pre-trained on image-based tasks retain significantly more image information compared to those trained only on contrastive learning tasks.</p>
               </div>
             </div>
             <div class="box">