Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
yarandex
on June 23, 2022
|
parent
|
context
|
favorite
| on:
YaLM-100B: Pretrained language model with 100B par...
Assuming running on CPU is memory-bandwidth limited, not CPU-limited, it should take about 200GB / (50GB/sec) = 4 seconds per character. Not too bad.
lostmsu
on June 23, 2022
[–]
That's per token. And you can generate quite a few per pass.
Consider applying for YC's Spring batch! Applications are open till Feb 11.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
