undefined
@evan I'm not sure how it could not be derivative. If the training data didn't slurp up all the repos on Github, it would not be able to generate something that looked an awful lot like a repo in Github.
If I trained a model on the collected literary works in the public domain and then asked it to produce a for loop written in Zig, no amount of prompting could make that happen.
On what grounds would that not be derivative?
@cthos That's kind of how it felt.
An anecdote to support your hypothesis.
So, From Jan 20 - Feb 20, I had the $20 subscription to Claude because the Verge and Verge adjacent folks came back from their holiday and were like "it's actually good."
I trusted this because while the Verge can be credulous about tech they're generally solid.
So, I try it. Mid trial Anthropic gives my account $50 dollars more in credits to trial Opus 4.6. I ended up hard setting it back to 4.5 because the only difference seemed to be churning through more tokens. The delta between quality and token usage was not visible.
Anyhow, the one month trial only served to make me more of a hater. Mostly because I got swindled out of $20 to confirm what I had already reasoned myself to.
@futurebird best is hard.
Most fulfilling: got through the process with a cold application and am really proud of that
Best quality of life: went through a process but there was a recruiter involved so that's like knowing a person
@futurebird best is hard.
Most fulfilling: got through the process with a cold application and am really proud of that
Best quality of life: went through a process but there was a recruiter involved so that's like knowing a person