Would be nice, but I was thinking of metrics that require “we’ve done the hard work of understanding our models and making them more reliable”, better neuron explanation seems more like it’s another smartness test.
Yeah, I agree it’s largely smartness, and I agree that it’d also be nice to have more non-smartness benchmarks—but I think an auto-interp-based thing would be a substantial improvement over current smartness benchmarks.
Maybe we should make fake datasets for this? Neurons often aren’t that interpretable and we’re still confused about SAE features a lot of the time. It would be nice to distinguish “can do autointerp | interpretable generating function of complexity x” from “can do autointerp”.
Would be nice, but I was thinking of metrics that require “we’ve done the hard work of understanding our models and making them more reliable”, better neuron explanation seems more like it’s another smartness test.
Yeah, I agree it’s largely smartness, and I agree that it’d also be nice to have more non-smartness benchmarks—but I think an auto-interp-based thing would be a substantial improvement over current smartness benchmarks.
Maybe we should make fake datasets for this? Neurons often aren’t that interpretable and we’re still confused about SAE features a lot of the time. It would be nice to distinguish “can do autointerp | interpretable generating function of complexity x” from “can do autointerp”.