Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
JimDabell
on March 14, 2023
|
parent
|
context
|
favorite
| on:
GPT-4
You can already use an LLM to train a smaller, more efficient LLM without significant loss in results.
canoebuilder
on March 14, 2023
[–]
Do you mean the output of a LLM as the training data for the new model? What is the specification for the prompts that generate the training data?
Any links with more info?
luxcem
on March 14, 2023
|
parent
[–]
There were just an article submitted few days ago about Alpaca, a LLM trained on GPT prompts:
https://news.ycombinator.com/item?id=35136624
canoebuilder
on March 15, 2023
|
root
|
parent
[–]
Thanks!
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
