Mamba & Jamba
First there was Mamba… now there is Jamba from AI21. This is a model that combines the best non-transformer goodness of Mamba with good ‘ol attention layers. This results in a highly performant and efficient model that AI21 has open sourced! We hear all about it (along with a variety of other LLM things) from AI21’s co-founder Yoav.
Discussion
Sign in or Join to comment or subscribe
Dylan Rund
2024-04-28T15:53:48Z ago
Hey Guys. I really enjoy this podcast so far as a new listener and an aspiring AI infrastructure guy. I was a little overwhelmed with the amount of episodes and did t want to start the beginning because how quickly information can change. Do you have any recommended episodes that are a must watch?