this post was submitted on 19 Jul 2023
390 points (95.8% liked)
Technology
58303 readers
10 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Can one suggest a good explation why? The model is trained and stays "as is". So why? Does opensi uses users rating (thumb up/down) for fine tuning or what?
AI trains on available content.
The new content since AI contai s a lot of AI - created content.
It's learning from its own lack of fully understood reality, which is degrading its own understanding of reality.
GPT is a pretrained model, "pretrained" is part of the name. It assumed to be static.
There is no gpt5, and gpt4 gets constant updates, so it's a bit of a misnomer at this point in its lifespan.
It's possible to apply a layer of fine-tuning "on top" of the base pretrained model. I'm sure OpenAI has been doing that a lot, and including ever more "don't run through puddles and splash pedestrians" restrictions that are making it harder and harder for the model to think.
They don’t want it to say dumb things, so they train it to say “I’m sorry, I cannot do that” to different prompts. This has been known to degrade the quality of the model for quite some time, so this is probably the likely reason.
more users means less computing power per user
Hm... Probably. I read something about chatgpt tricks in this area. Theoretically, this should impact web chat, not Impact API (where you pay for usage of the model).