Update model card: restore eos_token_id=None workaround, fix BERT mask_token, correct RNA prompt

README.md

@@ -19,7 +19,7 @@ GPT-2 medium (345M parameters) foundation model pre-trained on protein amino aci
 
 - **Model Type**: gpt2
 - **Data Type**: Protein
-- **Training Date**: 2026-04-13
+- **Training Date**: 2026-04-14
 
 ## Usage
 
@@ -39,7 +39,7 @@ with torch.no_grad():
         max_new_tokens=50,
         do_sample=True,
         temperature=0.8,
-        eos_token_id=tokenizer.eos_token_id,
+        eos_token_id=None,  # HF config.json has legacy eos_token_id=0; disable early stop
         pad_token_id=0,
     )
 print(tokenizer.decode(output_ids[0], skip_special_tokens=True))