Introducing Jamba: AI21's Groundbreaking SSM-Transformer Model

Good news! Jumping from Mamba state space model to Jamba! Perhaps attention is not all you need!

Caveat: I have not yet read the research paper on Mamba. It is on my to do list! 😊

Introducing Jamba: AI21's Groundbreaking SSM-Transformer Model Debuting the first production-grade Mamba-based model delivering best-in-class quality and performance.

Credits: Last Week in AI