test

docs/_pages/tutorial.md

@@ -3,6 +3,12 @@ title: "LLM2Vec Tutorial: Steps for transforming any decoder-only model into a t
 permalink: /tutorial/
 ---
 
+<script type="text/javascript"
+
+  src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.3/MathJax.js?config=TeX-AMS-MML_HTMLorMML">
+
+</script>
+
 LLM2Vec consists of 3 simple steps to transform decoder-only LLMs into text encoders: 1) enabling bidirectional attention, 2) training with masked next token prediction, and 3) unsupervised contrastive learning. After the LLM2Vec transformation, the model can be further fine-tuned with supervised data. Here, we provide a tutorial on how to use the LlaMA models.
 
 This tutorial will focus on the first two steps. After completing these steps, the model can be trained for unsupervised or supervised contrastive learning like any other encoder model.