Devinterp

What's Changed
* Add Basic Support for TPUs, refactor temp -> nbeta by jqhoogland in https://github.com/timaeus-research/devinterp/pull/81
* Will/Non-TPU dependent Integration with Aether by wz-ml in https://github.com/timaeus-research/devinterp/pull/92
* Quick fixes (use nbeta by default, though allow passing in temperature as a fallback as well) by wz-ml in https://github.com/timaeus-research/devinterp/pull/93


**Full Changelog**: https://github.com/timaeus-research/devinterp/compare/v1.0.0...v1.1.0

What's Changed
* All `estimate` functions now use `evaluate` instead of `criterion`, which can be constructed using the helper function `make_evaluate(criterion)`. A custom make_evaluate or evaluate_function allows preprocessing the data, making the package compatible with transformers. Example `evaluate` functions `evaluate_mse` and `evaluate_ce` can be imported from `utils.py`.
* Update estimate_learning_coeff functions by georgeyw in https://github.com/timaeus-research/devinterp/pull/77
* Add support for non-regression loss functions by jqhoogland in https://github.com/timaeus-research/devinterp/pull/80
* RLLC v1 by svwingerden in https://github.com/timaeus-research/devinterp/pull/74
* Add gradient accumulation for LLC estimation by georgeyw in https://github.com/timaeus-research/devinterp/pull/76
* Fixing two issues in MalaAcceptanceRate callback by rohan-hitchcock in https://github.com/timaeus-research/devinterp/pull/87
* Epsilon-Beta Sweeping & Plotting by wz-ml in https://github.com/timaeus-research/devinterp/pull/86

New Contributors
* rohan-hitchcock made their first contribution in https://github.com/timaeus-research/devinterp/pull/87
* wz-ml made their first contribution in https://github.com/timaeus-research/devinterp/pull/86

**Full Changelog**: https://github.com/timaeus-research/devinterp/compare/v0.2.2...v1.0.0

Removed an unnecessary warning, and fixed requirements to work with Python 3.8

**Full Changelog**: https://github.com/timaeus-research/devinterp/compare/v0.2.0...v0.2.2