Focus

This open-source model scores 95.4% on BIG-Bench Hard’s logic

Press Space to continue

This open-source model scores 95.4% on BIG-Bench Hard’s logic subset with just 3.66B parameters. TwIL-LM3-Pro is a model small enough to run locally on a laptop. Specialized post-training is super cool. You can get a ton of small models when you focus on specific tasks. https://x.com/svpino/status/210538030825… https://twitter.com/1242676339553026049/…

open on x · 2 views

Press Space to continue

This open-source model scores 95.4% on BIG-Bench Hard’s logic

Post

This open-source model scores 95.4% on BIG-Bench Hard’s logic