Update README.md

README.md

@@ -16,7 +16,7 @@ The model was fine-tuned on a subset of 96,780 sentences from the Ubertext datas
 Dataset [skypro1111/ubertext-2-news-verbalized](https://huggingface.co/datasets/skypro1111/ubertext-2-news-verbalized)
 
 ## Training Procedure
-The model underwent nearly 70,000 training steps, amounting to almost 2 epochs, to ensure thorough learning from the training dataset.
+The model underwent 70,000 training steps, which is almost 2 epochs, with further training the results degraded.
 
 ```python
 from transformers import MBartForConditionalGeneration, AutoTokenizer, Trainer, TrainingArguments