Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
michaelortega01
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
35 ms
·
1.
▲
by
michaelortega01
2y ago
At Predibase, we recently conducted 700+ fine-tuning experiments to benchmark the performance of popular open-source LLMs across 30 tasks and compared their results to GPT-4. 85% of the time they beat GPT-4. You can see the results here: h
2.
▲
by
michaelortega01
3y ago
Here's the blog: https://predibase.com/blog/lora-land-fine-tuned-open-source-...