The Liminal Learning Portal

@CMc_PhD Interesting that GPT-4 does so much better.

GPT-4. 9 out of 10 correct: Who are 10 prominent guests who have appeared on the Jim Rutt Show podcast? 1 / 1 As of my knowledge cutoff in September 2021, here are ten prominent guests who have appeared on the Jim Rutt Show podcast. The podcast features discussions on complex…

Got access to Bard. On my test queries it hallucinates like crazy. Worse than GPT3.5. Way worse than GPT-4. And rather prolix with its hallucinating. For example: only one (Rushkoff) is real:

Twitter has released the code for its recommendation algorithms. Anybody taken a look yet? https://t.co/I9X9qdRqpb

@b1daly And all that is a good thing.

@Dpyche The most wrecked I ever got on THC was cheffing up a batch of brownies that used a whole ounce of good 1980s vintage pot. The rest of the crew were out hiking and I ended up “sampling” 4 brownies. I was utterly stupefied. When my friends got back they found me sitting in the…

@parabanger Lid, love it! Haven’t heard that fine term since about 1974!

GPT3.5: gets 6 out of 10 right

@VladLiss No nausea that I recall even from mega doses. Munchies, yea.

@MashTunTimmy @RizomaSchool I do a quite tolerable House Of The Rising Sun. Probably peak quality in the 4 shots of bourbon range.