lol, yeah, that felt very appropriate ๐
Thank you for your kind words, my friend. It is an honor to be welcomed to your home, this beautiful land. I hope tomorrow brings more unity and kindness, and I will do my best to do my part to make that happen.
Happy Canada Day! Thank you for this beautiful post, it made me tear up a bit. I've made Canada my adopted home 14 years ago, and not a day goes by when I don't thank my fucking stars.
Okay, but can you let me know what it means after you get back?
From what my friend told me about Saudi Arabia, they have a similar mindset. With their extreme puritanical outlet, meeting women is near impossible. Being a top isn't considered being gay, but being a bottom is considered disgraceful.
The mental gymnastics, my god.
OMG, LilDumpy left another comment! ๐ธ
Wait, if it's double-blind, wouldn't the scientists be unaware of whether he got a placebo or the real drug? Although I suppose this could be after the study has concluded. I would bet that in real studies they record the side effects reported before checking to see if it was the placebo or not!
This is probably because of the autoregressive nature of LLMs, and is why "step by step" and "chain of thought" prompting work so well. GPT4 can only "see" up to the next token, and doesn't know how its own entire answer upfront.
If my guess is correct, GPT4 knew the probabilities of "Yes" or "No" were highest amongst possible tokens as it started generating the answer, but, it didn't really know the right answer until it got to the arithmetic calculation tokens (the 0.9 * 500 part). In this case it probably had a lot of training data to confirm the right value for 0.9 * 500.
I'm actually impressed it managed to correct course instead of confabulating!
You monster