31–40 of 226 posts
ChatGPT keeps track of context in a "discussion", but I found that alpaca 7B has no memory of what was said by the user or the model before, so each question gets answered with a blank slate; is that true, and if so, is this something that could be taught to the alpaca model as well?
* inject the running chat log into the prompt * inject the summary of the chat into the prompt
Wow sorry y'all, we didn't expect this to take off so quickly. Working on getting this scaled up now so everyone can play!
Wow sorry y'all, we didn't expect this to take off so quickly. Working on getting this scaled up now so everyone can play!
And we're back! Might be little slow while traffic is high but responses are coming through.
This is way too slow
how to explain being wet to a whale? To explain being wet to a whale, you could say something like, "It's like when you get out of the shower and you're all wet." Or, "It's like when you get out of the pool and you're all wet." Or, "It's like when you get out of the ocean and you're all wet." Or, "It's like when you get out of the lake and you're all wet." Or, "It's like when you get out of the rain and you're all we…
ChatGtp-4: Sure! Imagine telling a whale: "Hey, you know that weird feeling when you jump out of the water and the air tickles your skin? That's how we humans feel when we're wet, except we're not as good at making a big splash!"
how to explain being wet to a whale? To explain being wet to a whale, you could say something like, "It's like when you get out of the shower and you're all wet." Or, "It's like when you get out of the pool and you're all wet." Or, "It's like when you get out of the ocean and you're all wet." Or, "It's like when you get out of the lake and you're all wet." Or, "It's like when you get out of the rain and you're all we…
Oh you didn’t hear that? You must not be able to hear the high frequency.
ChatGPT keeps track of context in a "discussion", but I found that alpaca 7B has no memory of what was said by the user or the model before, so each question gets answered with a blank slate; is that true, and if so, is this something that could be taught to the alpaca model as well?
to add state (memory), u can either: * inject the running chat log into the prompt * inject the summary of the chat into the prompt
You can also fine tune the model to incorporate larger amounts of data, but that may be more expensive (and slower)
This kind of sounds like human short term and long term memory. Maybe “fine tuning” is analogous what happens to our memory when we sleep.
Coming soon: HN_LLaMA. Posts insightful, relevant, AI-generated HN comments to help boost your karma.
I'm eagerly waiting for the bot that automates the process of being "that guy" who copy-pastes the premise of every single article into chat-gpt and reporting what it says back.
-chatGPT