Update the correct AutoModel

README.md

@@ -3,9 +3,9 @@ This network has been fine-tuned for the task described in the paper *Topical Ch
 
 # Load the model
 ```python
-from transformers import AutoModel, AutoTokenizer
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
 tokenizer = AutoTokenizer.from_pretrained('dennlinger/roberta-cls-consec')
-model = AutoModel.from_pretrained('dennlinger/roberta-cls-consec')
+model = AutoModelForSequenceClassification.from_pretrained('dennlinger/roberta-cls-consec')
 ```
 
 # Input Format