docs: Update docs (#2)

* Added tokenlearn blogpost * Update * Update * Added gitignore * Removed site --------- Co-authored-by: stephantul <[email protected]>
MinishLab · Oct 29, 2024 · bef738f · bef738f
1 parent d4d0cf3
commit bef738f
Show file tree

Hide file tree

Showing 18 changed files with 3 additions and 3,612 deletions.
diff --git a/.gitignore b/.gitignore
@@ -0,0 +1,2 @@
+_site/*
+.sass-cache/*
diff --git a/_posts/2024-10-29-tokenlearn_blogpost.md b/_posts/2024-10-29-tokenlearn_blogpost.md
@@ -1,14 +1,12 @@
 ---
 layout: post
-title: Tokenlearn blogpost
+title: "POTION: bag of tricks leads to better models"
 categories: [Model2Vec]
 ---
 
 This blogpost describes the [Tokenlearn](https://github.com/MinishLab/tokenlearn) method, which is a method to pre-train Model2Vec models.
 
 
-# Bag of tricks leads to better models: POTION
-
 We've been brewing, concocting, distilling, and came up with a new distillation technique that leads to much better models, which we are now releasing under the name POTION. We open source all models, code, and data. 
 
 We're releasing three versions: a 64-dim (1.9M params), 128-dim (3.8M params), and 256-dim (7.6M params) model, all based on the same base model, which is, in turn, a bge-base distillation. All POTION models outperform all previous distillations in their size class, and should be considered to be drop-in replacements of our M2V_base_output model. potion-base-8M, in particular, even improves over our largest model, M2V_base_glove. potion-base-8M is better than any set of static embeddings we could find on any task, including glove, fasttext and specialized word embeddings.

diff --git a/_site/404.html b/_site/404.html
diff --git a/_site/about/index.html b/_site/about/index.html
diff --git a/_site/archive/index.html b/_site/archive/index.html