My average eventually made it to about 3900, and then stagnated between 3600-3900. I'm curious if this is universal behavior or not. I'm up to about 5k steps.
This idea is great in concept, and I think it's important to state that, but the GitHub Actions stuff is against TOS iirc + they will need to address that pretty quickly.
I believe this has the same issue as the last article that had these claims.
We can assume that Mythos was given a much less pointed prompt/was able to come up with these vulnerabilities without specificity, while smaller models like Opus/GPT 5.4 had to be given a specific area or hints about where the vulnerability lives.
I'm glad that Atari was willing to compromise at all. I'm happy with the updated response, and hope that it helps others understand the nuance of the situation. Anyone can still go download the main release from the official site.
It may say more about me than the person writing these type of README's, but if I see more than one or two emojis in a README, I immediately assume it was fully generated rather than written.
Very interesting. I would be curious to understand how granular these updates are being applied to CC + what might be causing things like this. I feel like I can notice a very small degradation but have compensated with more detailed prompts (which I think, perhaps naively, is offsetting this issue).
If I understand this correctly, upcoming Ca-derivations will fix this by making these situations expected, properly-handled cases rather than a weird bug? https://nixos.wiki/wiki/Ca-derivations
This is really incredible. I love easter eggs/hidden things in spectrograms. The implications of this are really cool, regardless of whether is it lossy or not.