Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
Ninjinka
6 days ago
|
parent
|
context
|
favorite
| on:
Learning to Reason with LLMs
Someone give this model an IQ test stat.
adverbly
6 days ago
|
next
[–]
You're kidding right? The tests they gave it are probably better tests than IQ tests at determining actually useful problem solving skills...
reply
Vecr
6 days ago
|
prev
[–]
It can't do large portions of the parts of an IQ test (not multi-modal). Otherwise I think it's essentially superhuman, modulo tokenization issues (please start running byte-by-byte or at least come up with a better tokenizer).
reply
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: