brady

like the nettspend of machine learning

[brady @bfloat8]

First ever blogpost!

[brady @bfloat8]

@our_decay are you going to see The Devils?

[brady @bfloat8]

@ManassehA06 @ojoshe @kleddamag @wand_125 @guzhou0806 @ctjlewis 17 next?

[brady @bfloat8]

does european csv taste better?

[brady @bfloat8]

this has more views and likes than the navier stokes announcement btw

[Jacob Coxon @hilbertspaess]

I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.

[brady @bfloat8]

they can't make me buy more storage, I will simply continue to compress the dataset each time it approaches using all my storage

[brady @bfloat8]

you will never be him.

[brady @bfloat8]

need to get used to reading VR200

[brady @bfloat8]

🤔

[brady @bfloat8]

https://t.co/JfJ1e2PBkN

[brady @bfloat8]

would really love to see how weight decay changes this plot

[brady @bfloat8]

god i love parquet files

[brady @bfloat8]

what ever happened to canon layers?

[brady @bfloat8]

@essenaccount https://t.co/VkDpW4Zbik

[typedfemale @typedfemale]

"are you memory bandwidth or compute bound?!?"
"memory bandwidth bound"
"so you *do* know the difference"

[brady @bfloat8]

@unixpickle https://t.co/S3GEP0wIxH

[brady @bfloat8]

@kalomaze crypto really is a plague

[brady @bfloat8]

its like a bicycle for your bowel

[brady @bfloat8]

if claude says "its worth being precise" one more time im going to lose it

[Charli @charli_xcx]

bladee best lyricist of all time and @itsthedare also agrees so yeah !