Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
pocketarc
on Sept 12, 2023
|
parent
|
context
|
favorite
| on:
Fine-tune your own Llama 2 to replace GPT-3.5/4
That is more than an order of magnitude better than my experience; I get around 2 t/s with similar hardware. I had also seen others reporting similar figures to mine so I assumed it was normal. Is there a secret to what you're doing?
ramesh31
on Sept 12, 2023
[–]
>Is there a secret to what you're doing?
Core speed and memory bandwidth matter a lot. This is on a Ryzen 7950 with DDR5.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: