# T5-base model create spelling mistake is summary
**URL:**
**Author:**  [@jiten-duhan](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fjiten-duhan)%5C
**Post date:** [September 14, 2020, 9:33am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ft5-base-model-create-spelling-mistake-is-summary%2F1116%2F1 "2020-09-14T09:33:26Z")
Hi,
I am using T5-base model for abstractive summarization, results are good but I am getting newly generated spelling mistakes in the summary which were not actually present in input text.
Can anyone tell me why these spelling mistakes occuring and how can I solve this?
---
**Author:**  [@Zack](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FZack)%5C
**Post date:** [October 24, 2020, 8:02pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ft5-base-model-create-spelling-mistake-is-summary%2F1116%2F2 "2020-10-24T20:02:57Z")
I think it’s due to your min\_output size, for example if you have forced the model to generate results at minimum more than 50 sequence, and somehow the prediction length predicted only 40 sequences, I think it will start to generate random tokens just to reach the 50 seq.
---
**Author:**  [@jiten-duhan](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fjiten-duhan)%5C
**Post date:** [November 5, 2020, 8:42am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ft5-base-model-create-spelling-mistake-is-summary%2F1116%2F3 "2020-11-05T08:42:40Z")
Hi Zack, hope you are doing well !!
Thank you for your reply.
Actually it is not generating random tokens, but it is misspelling them.
For e.g a word “productive” in input text is spelled as “priductive”