Update model card

README.md

@@ -88,9 +88,9 @@ tokenizer = AutoTokenizer.from_pretrained("lorenzocc/NeoBERTugues")
 fill_mask = pipeline("fill-mask", model=model, tokenizer=tokenizer)
 
 # Example usage
-result = fill_mask("O Brasil e um pais <mask>.")
+result = fill_mask("O Brasil é um país <mask>.")
 for r in result:
-    print(f"{r['token_str']}: {r['score']:.4f}")
+    print(f"{r['token_str']}: {r['score']:.1%}")
 ```
 
 ### Feature Extraction
@@ -115,20 +115,6 @@ masked_hidden = last_hidden * attention_mask.unsqueeze(-1)
 sentence_embedding = masked_hidden.sum(dim=1) / attention_mask.sum(dim=1, keepdim=True)
 ```
 
-### Fine-tuning for Classification
-
-```python
-from transformers import AutoModelForSequenceClassification, AutoTokenizer, Trainer, TrainingArguments
-
-model = AutoModelForSequenceClassification.from_pretrained(
-    "lorenzocc/NeoBERTugues",
-    num_labels=2
-)
-tokenizer = AutoTokenizer.from_pretrained("lorenzocc/NeoBERTugues")
-
-# Continue with your fine-tuning pipeline...
-```
-
 ## Model Architecture
 
 | Parameter                       | Value  |
@@ -164,7 +150,7 @@ tokenizer = AutoTokenizer.from_pretrained("lorenzocc/NeoBERTugues")
 If you use this model, please cite:
 
 ```bibtex
-@misc{cesconetto2025neobertugues,
+@misc{cesconetto2026neobertugues,
   author = {Cesconetto, Lorenzo},
   title = {NeoBERTugues: A Portuguese ModernBERT Model},
   year = {2026},