typo

index.html

@@ -1425,7 +1425,7 @@ <h3 id="benchmark_llama" class="">LLama2 Benchmark</h3>
 
 
 	<h3 id="benchmark_vit" class="">ViT Benchmark</h3>
-	<p>We evaluate the effectiveness of our quantization method on vision models as well. More specifically, we quantize various <a href="https://arxiv.org/abs/2103.00020">OpenCLIP</a> models from the <a href="https://arxiv.org/abs/2010.11929"> Visual Transformers (ViT)</a> family trained on the <a href="https://arxiv.org/abs/2210.08402"> LAION dataset.</a> Since Auto-GPTQ and Auto-AWQ calibration only work with text inputs, we can only evaluate against bitsandbytes by replacing all the linear layers inside the transformer blocks with their quantized versions. 
+	<p>We evaluate the effectiveness of our quantization method on vision models as well. More specifically, we quantize various <a href="https://arxiv.org/abs/2103.00020">OpenCLIP</a> models from the <a href="https://arxiv.org/abs/2010.11929"> Visual Transformers (ViT)</a> family trained on the <a href="https://arxiv.org/abs/2210.08402"> LAION dataset.</a> Since Auto-GPTQ and Auto-AWQ calibration only works with text inputs, we can only evaluate against bitsandbytes by replacing all the linear layers inside the transformer blocks with their quantized versions. 
 	</p>