Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
dauertewigkeit
on Feb 11, 2023
|
parent
|
context
|
favorite
| on:
How to train large models on many GPUs? (2021)
I tried that as well, but maybe I did not use it correctly. I did not see the full sharding that I was hoping for. I only saw results similiar to FSDP.
cma
on Feb 11, 2023
[–]
How about flexflow?
https://huggingface.co/transformers/v4.9.2/parallelism.html#...
Consider applying for YC's Spring batch! Applications are open till Feb 11.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: