I only have 16GB of RAM in my laptop, but I would love to run a powerful model like that even at 0.5 tokens/s. For the questions I am asking, such a speed is more than enough. If anyone has any suggestions, please let me know! I am not an expert on LLMs, just a humble mathematician.
Only tangentially related, but this is why I gave up on fountain pens. They are cool and elegant, far too cool to be a practical tool for me. I realized this at one time I was in Bonn; I was thinking deeply about a problem as I was holding this wonderful Sailor Pro Gear in my hand. At some point it slipped, fell on the desk and bent the nib sideways, a total disaster. That broke my stream of thought and ruined my day. Nah. Tools need to cater to me, not vice versa.
How on earth do you get people to talk in IRC? Am I missing something? Almost every channel I have joined is dead silent, and any time I tried starting a conversation I felt like I was asking Lebowski for my money.
The first appendix to the Common Core text for the English Language Arts contains some very interesting discussion of the role of advanced literacy outside the confines of academia. It is worth a read.
I was raised in a working-class family in Greece in the 1980s. We lived in what the average US person would describe as "squalor". Cockroaches crawling on our faces style.
My parents made only a few luxury expenses: encyclopedias for us children, and especially books about space and the cosmos. So please speak for yourself when talking about the interests of struggling people.
Perhaps. For now, one of my one-on-one tutoring sessions (in real analysis) this semester consisted mostly on un-teaching a student a bunch of wrong crap they "learned" from ChatGPT.