From 239e811bc71f4efa0789dee9b8d6cecb3f60b614 Mon Sep 17 00:00:00 2001
From: Daniel Han <danielhanchen@gmail.com>
Date: Mon, 14 Oct 2024 01:11:15 -0700
Subject: [PATCH] Updated Home (markdown)

---
 Home.md | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/Home.md b/Home.md
index 1a8801b..ef015c5 100644
--- a/Home.md
+++ b/Home.md
@@ -263,6 +263,16 @@ tokenizer = get_chat_template(
 )
 ```
 
+### Adding new tokens
+Unsloth has a function called `add_new_tokens` which allows you to add new tokens to your finetune. For example if you want to add `<CHARACTER_1>`, `<THINKING>` and `<SCRATCH_PAD>` we can do the following:
+```python
+model, tokenizer = FastLanguageModel.from_pretrained(...)
+from unsloth import add_new_tokens
+add_new_tokens(model, tokenizer, new_tokens = ["<CHARACTER_1>", "<THINKING>", "<SCRATCH_PAD>"])
+model = FastLanguageModel.get_peft_model(...)
+```
+Note - you MUST always call `add_new_tokens` before `FastLanguageModel.get_peft_model`!
+
 ### 2x Faster Inference
 Unsloth supports natively 2x faster inference. All QLoRA, LoRA and non LoRA inference paths are 2x faster. This requires no change of code or any new dependencies.
 ```python