Bharat Bheesetti
@bharatbheesetti.bsky.social
www.bharatbheesetti.com
Product builder. Transhumanist.
Love ML, film, sci fi, Man Utd and vidya.
Currently building ML data platforms at Telus International.
Product builder. Transhumanist.
Love ML, film, sci fi, Man Utd and vidya.
Currently building ML data platforms at Telus International.
First time going to a concert where I don't know any of the songs.
December 6, 2024 at 12:32 PM
First time going to a concert where I don't know any of the songs.
We need more people like Cory Doctorow in this world
November 27, 2024 at 2:09 PM
We need more people like Cory Doctorow in this world
This is excellent. Benchmarks are actually super hard to get right in a way that keeps them being right. Any public test worth its salt will be duplicated and rendered worthless sooner rather than later.
I wrote some thoughts on how to build good LM benchmarks: ofir.io/How-to-Build...
How to Build Good Language Modeling Benchmarks
Building benchmarks is important because they shine a spotlight on the weaknesses of existing language models and so can guide the community on how to improve them.
ofir.io
November 26, 2024 at 1:26 PM
This is excellent. Benchmarks are actually super hard to get right in a way that keeps them being right. Any public test worth its salt will be duplicated and rendered worthless sooner rather than later.
Barnes & Noble wins if we do this, but I'd say it's still worth it. Life truly is too short.
Life is short. It’s ok to stop reading something you’re not digging.
No cap.
No cap.
November 21, 2024 at 1:11 PM
Barnes & Noble wins if we do this, but I'd say it's still worth it. Life truly is too short.
Blueberries are the apex fruit
November 20, 2024 at 7:31 PM
Blueberries are the apex fruit
Reposted by Bharat Bheesetti
The Llama 3.2 1B and 3B models are my favorite LLMs -- small but very capable.
If you want to understand how the architectures look like under the hood, I implemented them from scratch (one of the best ways to learn): github.com/rasbt/LLMs-f...
If you want to understand how the architectures look like under the hood, I implemented them from scratch (one of the best ways to learn): github.com/rasbt/LLMs-f...
November 20, 2024 at 8:33 AM
The Llama 3.2 1B and 3B models are my favorite LLMs -- small but very capable.
If you want to understand how the architectures look like under the hood, I implemented them from scratch (one of the best ways to learn): github.com/rasbt/LLMs-f...
If you want to understand how the architectures look like under the hood, I implemented them from scratch (one of the best ways to learn): github.com/rasbt/LLMs-f...
Arcane is just fire from the first moment. Phenomenal show.
November 20, 2024 at 6:57 PM
Arcane is just fire from the first moment. Phenomenal show.
Bluesky gives me that Mastodon vibe. Nice.
November 20, 2024 at 4:22 PM
Bluesky gives me that Mastodon vibe. Nice.
Talk about starting from zero.
November 20, 2024 at 4:18 PM
Talk about starting from zero.