transformers can take maximum of 512 token at a time. it is a architecture constraint

Hey awsome article. One question. Why do you limit the text to 512 terms?
1
Marco
Subir Verma
·Follow
Nov 4, 2021
--
transformers can take maximum of 512 token at a time. it is a architecture constraint
--
--
Written by Subir Verma349 Followers
·108 Following
Lead Data Scientist at Freshworks (https://www.linkedin.com/in/sbrvrm/)
Responses (1)
Help
Status
About
Careers
Press
Blog
Privacy
Terms
Text to speech
Teams