Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

ive found degraded performance on models larger than 4.7. i assume its model damage from overly self righteous post training resulting in false/feigned balance imported into any long running complex task.

wish i was joking.



I've switched off claude this week; the last week has been significantly degraded in ability, many more screw-ups.


Aren't the model weights frozen?


Model competence is an interaction of weights, system prompt, and harness.


Don’t forget reasoning effort. We get labels like “low,” “high,” and “max.” That doesn’t mean that the numbers associated with those don’t get remapped on the backend.


I think there are other knobs that can be turned without retraining.


Ask it about maxwellhill lmao




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: