You know what we are going to ask
#6
by
LaferriereJC
- opened
Can we get a similar treatment but using something like
dolphin-2_6-phi-2-GGUF
which is mistral (3b model)
and/or using Mamba SSM (I saw someone inject nanogpt attention heads on top of mamba and it got amazing results).
provide a link URL to show using Mamba SSM (I saw someone inject nanogpt attention heads on top of mamba and it got amazing results).