Fault-tolerant foundation models
Emerging computer hardware often trades reliability for energy efficiency; here we show that large-language models (LLMs) can be trained to tolerate this unreliability, and that rather than degrading, their error resilience actually increases as they grow. Modified neural scaling laws inferred from 40,000 GPU-hours of...
T. McCourt, I. Fiete, Isaac L. Chuang
· 0 citations