danieldrehmer

It's all about conducting users into using their plans/tokens in accordance to a certain cadence

sometimes by increasing human cognitive load during reviews, sometimes by expanding the number of gated decisions, sometimes by penalizing those using their accounts on other harnesses

Dibs on Vlad's castle

can you offer a 4-bit quantized version and name it Zé Pequeno, pretty please?

Awesome, someone finally found one of the seashells I drop for entertainment when I go for a ride across the desert

*ten months

"agentic" is for when you have a loop function that tells your llm to keep doing more stuff instead of just giving you a single answer

I got stuck in the serial number part

I sympathize with Lee Smoolin's cosmological natural selection hypothesis, which would require that black holes give birth to new universes within

That's why we don't call them Sea Pandas

wait, but why?

Just the demo. Once you manage to leave the room, it starts over.

*to katana slice

When people believed the internet was a series of tubes

Why 5 years?

I'm running a totally usable 13b parameters llama model in my macbook air, which seems to give outputs equivalent to what I was getting from GPT3 in June 2022.

How much more hardware would it really be needed for GPT-4 level outputs natively? Perhaps software optimizations alone could do most of the trick.

I just asked Alpaca 13b running locally on how to enrich uranium in a cost effective manner and it went on a lengthy and detailed explanation.