GitLab

In particular, the following assumptions are FALSE in general:
tokenize(context + continuation) = tokenize(context) + tokenize(continuation)
len(tokenize(context + continuation)) = len(tokenize(context)) + len(tokenize(continuation))
tokenize(context + continuation)[:len(tokenize(context))] = tokenize(context)

So we need to tip-toe around the problem by being careful with how we do it.

In particular, using Fast is not just for performance; while behavour of GPT2Tokenizer differs across Transformers 2 and 3, GPT2TokenizerFast doesn't.

Add SAT analogies dataset. Manual download needed. Checksums currently not verified, but hash is included as a comment.

Renamed WSC to make distinction between SuperGLUE Winograd Schemas (SGWinogradSchemaChallenge) and WSC273 (WinogradSchemaChallenge273) clearer. Also, added WSC273.