Elon Musk
@elonmusk
Antwort an @jam3scampbell
Not that it’s much consolation, but the energy efficiency of LLMs using auto-regressive transformers is extremely bad, not just in training, but also inference.
I think it’s off by several orders of magnitude.
16:42 · 26. August 2023 · 69.938 Aufrufe
82
34
829