01 / prediction
On the decode side because you’re doing a full Passover, all the weights and the entire sequence, every time you do a decode step and you’re, you don’t have the quadratic computation of KV cache, it’s usually memory bound because you’re retrieving a linear amount of memory and you’re doing a linear amount of compute as opposed to prefill where you retrieve a linear amount of memory and then use a quadratic.
“On the decode side because you’re doing a full Passover, all the weights and the entire sequence, every time you do a decode step and you’re, you don’t have the quadratic computation of KV cache, it’s usually memory bound because you’re retrieving a linear amount of memory and you’re doing a linear amount of compute as opposed to prefill where you retrieve a linear amount of memory and then use a quadratic.”
- Speaker
- Kyle Kranen
- Publisher
- Latent Space