GitLab

fix translation scoring format

Implement SQuADv2 evaluation

# Conflicts:
#	lm_eval/tasks/squad.py

Add `Random` argument to `write_out` `fewshot_context`

Remove unused imports and format imports

Adding hendrycksTest (https://arxiv.org/abs/2009.03300)