3 ms·
Is it reasonable to expect that some companies out there are training LLMs powered by Hyena instead of attention? Like a GPT-3 level size to bet on ultra-long c
by alpark3 3y ago
Is it reasonable to expect that some companies out there are training LLMs powered by Hyena instead of attention? Like a GPT-3 level size to bet on ultra-long context length?
- theGnuMe 3y agoYes.