Comment on TransMLA: Multi-head latent attention is all you needComments−wiz21c1yNot quite related, but do the mamba models gain ground ?Answering my own question: https://www.reddit.com/r/MachineLearning/comments/1hpg91o/d_...
Comments
Not quite related, but do the mamba models gain ground ?
Answering my own question: https://www.reddit.com/r/MachineLearning/comments/1hpg91o/d_...