Describe the bug
When setting a prompt for generation and then making the tab inactive i.e. moving to another tab and browsing, more often than not, sillytavern receives the output from whatever model i was using but simply doesn't update teh UI with it. so when i return to the tavernAI UI in teh firefox browser, another generation gets requested for no reason so i have to wait again for another generation with no way to recover the old wasted generation.
Additional contect
Firefox browser
windows 10
silly tavern latest
using openAI proxies (any proxy has this problem. even turbo or gpt4)
Describe the bug When setting a prompt for generation and then making the tab inactive i.e. moving to another tab and browsing, more often than not, sillytavern receives the output from whatever model i was using but simply doesn't update teh UI with it. so when i return to the tavernAI UI in teh firefox browser, another generation gets requested for no reason so i have to wait again for another generation with no way to recover the old wasted generation.
Additional contect Firefox browser windows 10 silly tavern latest using openAI proxies (any proxy has this problem. even turbo or gpt4)