Tweets
2024-05-01T02:57:01+00:00 | 🔗
The only practical solution available, however flawed. Crucial details hidden though. Divestment doesn’t do anything. https://t.co/vkbsXcPeY7
2024-05-01T02:38:37+00:00 | 🔗
RT @nateobrienn: I’m fully convinced that this is the best part of the US to live in https://t.co/xUNe8zpdo7
2024-04-30T20:58:29+00:00 | 🔗
@jxmnop @init_malachi You are looking for an autoencoder. Diffusion models need a decoder, regardless.
2024-04-30T20:53:01+00:00 | 🔗
All the context and retrieval won't help if what you are looking for was never there is the first place.
2024-04-29T19:43:22+00:00 | 🔗
Some concept art for what a LLM workstation might look like https://t.co/Wwa0DyHoJ4
2024-04-29T17:31:14+00:00 | 🔗
RT @StephenESachs: 1/ It's nice to see my thread mentioned in @thecrimson last week, and again in this morning's story—but I wish they'd re…
2024-04-29T15:16:56+00:00 | 🔗
Bioweapons are not in an AI's training set, it's ridiculous. AI can't even fold laundry, and somehow it's suppose to miraculously walk someone through a bioweapon synthesis? The best labs have lots of little tips and tricks and secrets all squirreled away that AI has no access
2024-04-28T13:49:16+00:00 | 🔗
RT @DavidGQuaid: After people pointed out similarities between search results for Perplexity and Google in comments on X and Reddit, Sriniv…
2024-04-27T23:40:15+00:00 | 🔗
RT @teortaxesTex: The Synthetic Data Solves ASI pill is a bit awkward: - Our AI plateaues ≈on intelligence level of expert humans because i…