anematode

5 ms·

31.
by anematode 3mo ago
> You can continue to enjoy the books and articles etc Except for the books that Anthropic bought and destroyed? There's also the cultural displacement element: while yes, some people (including me) would seek out original content,
32.
by anematode 3mo ago
They're stealing, eh?
33.
by anematode 3mo ago
Greediest, perhaps?
34.
by anematode 3mo ago
There are indeed scrapers which use tens of thousands of distinct IPs, and so rate limiting them isn't a solution.
35.
by anematode 3mo ago
Should be `dogs[0].bark()`
36.
by anematode 3mo ago
It also contextualizes the urgency of their attempts at regulatory capture. Once Chinese models have the same capabilities, there's almost no reason not to use them.
37.
by anematode 3mo ago
It's fantastic to publish these. Otherwise Anthropic's commitment to "AI safety" is Emperor Dario's new clothes, its failure unacknowledged. An easy alternative is to switch to a Chinese open-weights model, adjust y
38.
by anematode 3mo ago
I know (via probing these models) that some of my work is in the training data. My mailbox is open.
39.
by anematode 3mo ago
Yup, it's hard to take seriously any complaint about "stealing" Anthropic's services, when their entire business is based on massive theft.
40.
by anematode 4mo ago
It's obviously an error, but I'm going to assume you are larping and move on
41.
by anematode 4mo ago
What?
42.
by anematode 4mo ago
The honest answer:
43.
by anematode 4mo ago
Utter dysfunction. I ripped out bun from my projects after the vibe Rust rewrite, but seems like the problem has existed much longer...
44.
by anematode 4mo ago
Why the snark? wasmtime is a (pretty popular) Rust project which uses a JIT, demonstrating that it's not incompatible with Rust. Obviously a proper VM wouldn't depend on wasmtime, but implement its own JIT and paraphernalia.
45.
by anematode 4mo ago
wasmtime exists
46.
by anematode 4mo ago
I'd actually love to see a relatively high-performance (i.e., including a decent JIT) runtime for a dynamic language that's written in Rust. There's a lot of implementations like Rust Python, the Boa JS engine, etc. that are
47.
by anematode 4mo ago
Thanks, fixed
48.
by anematode 4mo ago
This is terrifying. Evidently based on prior art by Mr. Pizlo – indeed, where's the acknowledgement of that?? (edit: I missed it) – but I'm assuming that was never translated into code. I love the idea of experimentation and innov
49.
by anematode 4mo ago
Thanks for the link, a good read
50.
by anematode 4mo ago
One thing that stuck out to me is that deals with a lot more data formats, in particular, low-precision formats like FP4, FP6 and FP8. Manipulating those formats can take a lot of annoying effort; in general, x86 (until AVX-512, at least) h
51.
by anematode 4mo ago
Context matters; how many of these images in the training data are taken from shock websites, and therefore associated with misanthropic commentary, versus legitimate sources like medical journals or historical pictures? Based on the sample
52.
by anematode 4mo ago
Legitimate criticism of the author's presentation aside, I'm quite disappointed by how many commenters here are justifying the model's output. I guess there's a lot of misanthropy and nihilism here? It's one thing t
53.
by anematode 4mo ago
This is far too simplistic. Some things just don't belong in the training data. Along similar lines, Grok was found to generate images of child sexual abuse: https://www.bbc.com/news/articles/cvg1mzlryxeo
54.
by anematode 4mo ago
Indeed, learning about mitochondria is more innocuous...
55.
by anematode 4mo ago
Agreed. Moreover, the authors of copyright law could never have anticipated this type and scale of abuse. Maybe the companies are legally in the right, maybe not, but that's irrelevant for the question of whether it's ethical. The
56.
by anematode 4mo ago
By writing a not-identical, but valid, solution? Any modestly complex engineering problem has many solutions. This is an obvious example of why LLM training is so different than human learning.
57.
by anematode 4mo ago
> memorization of upstream fixes from training data At least now we have up-to-date evidence on their laundering, and the fact that regurgitation absolutely still happens.
58.
by anematode 4mo ago
For cyberattacks especially, where things are often roughly interchangeable, I wonder if one could construct a harness where a "weaker" model asks questions that obfuscate the end purpose, but whose answers are still useful, and s
59.
by anematode 4mo ago
Maybe Dario should have cited The Lorax .
60.
by anematode 4mo ago
I'm not normally a fan of ex post facto law, but...

More ›