Update README.md

README.md

@@ -130,7 +130,7 @@ with torch.no_grad():
             model_input[k] = torch.tensor(batch[k]).to('cuda')
 
         output_ids = lama2dictionary.generate(**model_input,
-                                       max_length = max_length * batch_size,
+                                       max_length = max_length,
                                        forced_eos_token_id = eos_tokens,
                                        max_time = max_time * batch_size,
                                        eos_token_id = eos_tokens,