alt Hacker NewsThe activation capping effect on LLM behavior is available in this paper:
https://www.anthropic.com/research/assistant-axis
The estimated energy consumption versus error rate is likely projected from agent test and hidden-agent coverage.
You are correct, in that such a big number likely includes large errors itself given models change daily. =3
ok, your quote was over generalized, you meant "current LLM need..." and not "any conceivable LLM"
although the word "energy" does not appear on that page, not sure where you get the galaxy energy consumption from