Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
carlita_express
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
3 ms
·
1.
▲
by
carlita_express
1y ago
I suppose that is the question...
2.
▲
by
carlita_express
1y ago
I am aware of that, like I said: > (Or at least because O(log) increases in model performance became unreasonably costly?) But, yes, I left implicit in my comment that the trend might be “fleeting” because of its impracticality. RL is on
3.
▲
by
carlita_express
1y ago
> we’ve observed that large-scale reinforcement learning exhibits the same “more compute = better performance” trend observed in GPT‑series pretraining. Didn’t the pivot to RL from pretraining happen because the scaling “law” didn’t deli
4.
▲
by
carlita_express
5y ago
iOS has had this functionality for a while now, but android is a lot more diverse
5.
▲
by
carlita_express
6y ago
Most “public” school teachers have experience recognizing the symptoms of ADHD that parents don’t have.
6.
▲
by
carlita_express
6y ago
Anecdotally, I was homeschooled and experienced all of these above, in addition to my ADHD which was ignored by my parents, who didn't have enough experience to diagnose common symptoms.
7.
▲
by
carlita_express
6y ago
On this subject, there is a subreddit ( https://old.reddit.com/r/homeschoolrecovery ) dedicated to those trying to mentally/socially recover from their experiences of home/un-schooling, which can highlight so