Added architecture reminder in code-sample

README.md

@@ -20,7 +20,6 @@ The Tiny-Toxic-Detector achieves an impressive 90.26% on the Toxigen benchmark a
 | --------------------------------- | ----------------- | ----------- | ---------- |
 | lmsys/toxicchat-t5-large-v1.0     | 738M              | 72.67       | 88.82      |
 | s-nlp/roberta toxicity classifier | 124M              | 88.41       | **94.92**  |
-| unitary/toxic-bert                | 109M              | 49.50       | 89.70      |
 | mohsenfayyaz/toxicity-classifier  | 109M              | 81.50       | 83.31      |
 | martin-ha/toxic-comment-model     | 67M               | 68.02       | 91.56      |
 | **Tiny-toxic-detector**           | **2M**            | **90.26**   | 87.34      |
@@ -194,6 +193,7 @@ if __name__ == "__main__":
 
 Please note that to predict toxicity you can use the following example:
 ```python
+# Define architecture before this!
 inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=128, padding="max_length").to(device)
 if "token_type_ids" in inputs:
   del inputs["token_type_ids"]