Susan Zhang

I'm having a similar issue, running on an EC2 instance with Ubuntu 18.04 as host, though I'm not sure I'm doing anything correctly with netplan (still trying to figure out...

This seems right - here's a paper trail of directionally moving evals to streaming lm task: https://github.com/fairinternal/metaseq-internal/issues/54 I can't speak to the result of evals though, but replacing language modeling...

This seems unrelated to this task - please open a separate issue tracking. For reference, we are using version 2.1.1 for omegaconf (should pin this version here too).