cmpatino HF Staff commited on
Commit
4bed84f
·
1 Parent(s): b9a88c9

Import HTML embeds

Browse files
Files changed (1) hide show
  1. app/src/content/article.mdx +2 -0
app/src/content/article.mdx CHANGED
@@ -48,6 +48,8 @@ tableOfContentsAutoCollapse: true
48
  pdfProOnly: false
49
  ---
50
 
 
 
51
 
52
  On-policy distillation is a highly effective strategy for compressing LLMs, as recently highlighted by [Thinking Machines' excellent blog post.](https://thinkingmachines.ai/blog/on-policy-distillation/) The technique trains a small "student" model by transferring knowledge from a high-performing "teacher" model's probability distribution. This allows the student to emulate the teacher's task performance, while significantly reducing size and latency.
53
 
 
48
  pdfProOnly: false
49
  ---
50
 
51
+ import HtmlEmbed from '../components/HtmlEmbed.astro'
52
+
53
 
54
  On-policy distillation is a highly effective strategy for compressing LLMs, as recently highlighted by [Thinking Machines' excellent blog post.](https://thinkingmachines.ai/blog/on-policy-distillation/) The technique trains a small "student" model by transferring knowledge from a high-performing "teacher" model's probability distribution. This allows the student to emulate the teacher's task performance, while significantly reducing size and latency.
55