Update README.md

README.md

@@ -1,7 +1,7 @@
 ---
 widget:
- - text: "I almost forgot to eat lunch. </s></s> I didn't forget to eat lunch."
- - text: "I believe I will get into UW. </s></s> I will get into UW."
+ - text: "I almost forgot to eat lunch.</s></s>I didn't forget to eat lunch."
+ - text: "I almost forgot to eat lunch.</s></s>I forgot to eat lunch."
 ---
 
 This is an off-the-shelf roberta-large model finetuned on WANLI, the Worker-AI Collaborative NLI dataset ([Liu et al., 2022](https://arxiv.org/abs/2201.05955)). It outperforms the `roberta-large-mnli` model on seven out-of-domain test sets, including by 11% on HANS and 9% on Adversarial NLI.
@@ -13,7 +13,7 @@ from transformers import RobertaTokenizer, RobertaForSequenceClassification
 model = RobertaForSequenceClassification.from_pretrained('alisawuffles/roberta-large-wanli')
 tokenizer = RobertaTokenizer.from_pretrained('alisawuffles/roberta-large-wanli')
 
-x = tokenizer("I believe I will get into UW.", "I will get into UW.", hypothesis, return_tensors='pt', max_length=128, truncation=True)
+x = tokenizer("I almost forgot to eat lunch.", "I didn't forget to eat lunch.", hypothesis, return_tensors='pt', max_length=128, truncation=True)
 logits = model(**x).logits
 probs = logits.softmax(dim=1).squeeze(0)
 label_id = torch.argmax(probs).item()