Update README.md

README.md

@@ -15,7 +15,7 @@ This repo contains the code to apply supervised SAEs on LLMs. With this, LLMs ca
 # Usage
 
 Load the model weights from HuggingFace:
-```python3
+```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
 SCAR = AutoModelForCausalLM.from_pretrained(
@@ -26,7 +26,7 @@ SCAR = AutoModelForCausalLM.from_pretrained(
 
 The model loaded model is based on LLama3-8B base. So we can use the tokenizer from it:
 
-```python3
+```python
 tokenizer = AutoTokenizer.from_pretrained(
         "meta-llama/Meta-Llama-3-8B", padding_side="left"
     )
@@ -36,7 +36,7 @@ inputs = tokenizer(text, return_tensors="pt", padding=True)
 ```
 
 To modify the latent feature $h_0$ (`SCAR.hook.mod_features = 0`) of the SAE do the following:
-```python3 
+```python
 SCAR.hook.mod_features = 0
 SCAR.hook.mod_scaling = -100.0
 output = SCAR.generate(