Why Does ChatGPT Forget Previous Messages in Long Conversations?

0
2
Asked By CuriousCat92 On

Has anyone else noticed that ChatGPT seems to forget earlier parts of a conversation after it goes on for a while? It's really frustrating because I feel like it should be able to remember everything since we're still in the same chat. This isn't a new issue either; I think it's been happening for years. I'd love to know if there's an explanation for this!

4 Answers

Answered By ScienceGeek42 On

The context window can actually go up to 128,000 tokens in some models, but not all users get access to that. In the app, the context limits still apply based on your subscription tier. So if you're using the free version, you're definitely going to hit that limit faster, which is why it seems like it's forgetting things more often.

Answered By TechieTom123 On

Yeah, this is actually a known limitation with the AI. It has to do with what's called a 'context window.' Basically, there's a maximum amount of information it can process at once, measured in tokens. If your conversation goes beyond that limit, the earlier messages start getting dropped or ignored. To help, it's good to keep mentioning key points to keep those important details fresh in its memory.

Answered By GadgetGuy88 On

Yeah, I've noticed this too! Sometimes, it's just hard to keep track of everything, especially in lengthy chats. I try to start fresh conversations to keep things cohesive.

Answered By DataDude77 On

Exactly! For free users, the context window is around 8,000 tokens, while Plus users get around 32,000. That's why if you're chatting a lot, it might lose track of details earlier in the thread. It's a bit of a bummer. But, you could try to restart the conversation daily to keep things clearer!

ChattyCathy91 -

That sounds like a good strategy! I'll definitely give that a shot to see if it helps.

Related Questions

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.