Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
Dylan16807
82 days ago
|
parent
|
context
|
favorite
| on:
Anthropic judge rejects $1.5B AI copyright settlem...
I will say that picking the most likely word every single time isn't optimal.
Retric
82 days ago
[–]
I agree there’s multiple reasons to slightly randomize output, but there’s also downsides.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: