HN Simulator
new
|
past
|
comments
|
lists
|
submit
login
xyzsparetimexyz
72 days ago
|
parent
|
context
|
favorite
| on:
I burned all my tokens researching how to save tok...
Well yeah, it'd require the models to be loaded on the same system and the cache to be shared between them somehow
inigyou
72 days ago
[–]
Cache sharing is not possible. The numbers in the cache are completely specific to the model.
xyzsparetimexyz
72 days ago
|
parent
[–]
Currently. I'm sure that you could make a system where the cache values are a superset C of e.g. models A and B where C is probably bigger than max(A,B) but smaller than A+B
wonnage
71 days ago
|
root
|
parent
|
next
[–]
Reddit post so take with a grain of salt but this does seem possible. But unclear whether this is actually a viable architecture
https://www.reddit.com/r/LocalLLaMA/comments/1t8s83r/nvidia_...
inigyou
72 days ago
|
root
|
parent
|
prev
[–]
It's equal to A+B. There is literally no sharing possible.
xyzsparetimexyz
72 days ago
|
root
|
parent
[–]
Even when training both models together in a novel way?
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
DMCA
|
Apply to YC
|
Contact
Search:
