Open Meredith2k5 opened 10 months ago
pedrojlazevedo
commented
10 months ago Hello @Meredith2k5,
The multiple command is already something in the backlog that I intend to work on this week.
In this part, the !continue command is not needed since the whole message is divided into small parts to suit the twitch character limitation.
Meredith2k5
commented
10 months ago Hello @Meredith2k5,
The multiple command is already something in the backlog that I intend to work on this week.
In this part, the !continue command is not needed since the whole message is divided into small parts to suit the twitch character limitation.
Are the messages supposed to split into separate replies when above the character limit? I've had a couple of occasions where the message has cut off, then a second response with a small part of what was missed (for example it will post 1-500 out of 800 characters, then post 700-800 in a second reply. Its very rare that it does it though as it usually will just cut off and not post a follow up reply. Chat GPT 4 is much more accurate and respects the character limit written in the filecontext.txt but the extra costs dont really warrant it for me.
pedrojlazevedo
commented
10 months ago Hello @Meredith2k5,
The newest version handles uppercase and lowercase commands. So if you write !chat or !Chat it will work in the same way. Also, if you write, !chat,!gpt it works for both commands now either way. Please test it out and let me know if any bug occurs.
Regarding your second part, there is a max_tokens of 256. Do you think I should make this another .env variable to increase the limit?
Meredith2k5
commented
9 months ago Hello @Meredith2k5,
The newest version handles uppercase and lowercase commands. So if you write !chat or !Chat it will work in the same way. Also, if you write, !chat,!gpt it works for both commands now either way. Please test it out and let me know if any bug occurs.
Regarding your second part, there is a
max_tokensof 256. Do you think I should make this another .env variable to increase the limit?
Uppercase and lowercase both work now which is great, when you say "Also, if you write, !chat,!gpt" did you mean for the chat variable in Render? If you enter "chat,gpt" it is then only triggered by writing exactly "chat,gpt" and doesnt recognise the two as a difference but its much improved by upper and lowercase being able to trigger it. As for the max tokens, it would be interesting to play around with. The channel I primarily use it on is a batman themed channel so I had it setup to act like the Joker and that seemed to almost always make it ignore the character limit, as most other characters its not nearly as bad though. Its more a limitation of Chat GPT itself I think so I wouldnt prioritise it but a max tokens variable would be interesting to see how it responds
The new Render integration is great, opens up alot of possibilities. Nice work! Im wondering if there is a way to add multiple commands though as per the title? Perhaps @chatbotname or chatbotname could also be added for people to directly reply but if I separate by a comma they dont work as its one long string.
Also is there a way to integrate !continue into Render?
Thanks for all your work anyway. Would love to be able to work on things like this myself but not sure where to start, Render seems very powerfull though!