Transformer-lens

New minor release with a variety of improvements relating to testing, documentation and development. Transition from torchtyping to jaxtyping is one the most significant changes.

What's Changed
* Replace torchtyping with jaxtyping by dkamm in https://github.com/neelnanda-io/TransformerLens/pull/171
* Run poetry lock by rusheb in https://github.com/neelnanda-io/TransformerLens/pull/178
* Add `verbose` flag to disable tqdm on `model.generate(...)` by afspies in https://github.com/neelnanda-io/TransformerLens/pull/185
* Make `tracr` plot show outside on colab by ArthurConmy in https://github.com/neelnanda-io/TransformerLens/pull/184
* Add positional_embedding_type to model properties table by ckkissane in https://github.com/neelnanda-io/TransformerLens/pull/176
* Run `poetry lock --check` in CI by rusheb in https://github.com/neelnanda-io/TransformerLens/pull/182
* Slice: doc and tests by Xmaster6y in https://github.com/neelnanda-io/TransformerLens/pull/166
* Separate tests into unit and acceptance tests by rusheb in https://github.com/neelnanda-io/TransformerLens/pull/191
* Grokking demo by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/193
* Configure coverage reports to measure branch coverage by rusheb in https://github.com/neelnanda-io/TransformerLens/pull/192
* Silence DeprecationWarning for distutils by rusheb in https://github.com/neelnanda-io/TransformerLens/pull/187
* Clone pos embed by slavachalnev in https://github.com/neelnanda-io/TransformerLens/pull/194
* Add pos embed hook tests by slavachalnev in https://github.com/neelnanda-io/TransformerLens/pull/196
* Fix test command in Readme by valedan in https://github.com/neelnanda-io/TransformerLens/pull/197
* Test constructor of FactoredMatrix by rusheb in https://github.com/neelnanda-io/TransformerLens/pull/188
* Add helper for logit attribution by dkamm in https://github.com/neelnanda-io/TransformerLens/pull/135
* issue and pr templates by jbloomAus in https://github.com/neelnanda-io/TransformerLens/pull/203

New Contributors
* Xmaster6y made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/166
* slavachalnev made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/194
* valedan made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/197

**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.2...v1.2.1

What's Changed
* Test cache names by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/169
* Implement from_pretrained_no_processing by lukasberglund in https://github.com/neelnanda-io/TransformerLens/pull/161
* Minor Readme Changes by jbloomAus in https://github.com/neelnanda-io/TransformerLens/pull/170
* Add QKV split by ArthurConmy in https://github.com/neelnanda-io/TransformerLens/pull/158
* Add gotcha to HookedTransformer.to_str_tokens docstring by rusheb in https://github.com/neelnanda-io/TransformerLens/pull/173
* Test coverage reports by jbloomAus in https://github.com/neelnanda-io/TransformerLens/pull/172
* Sphinx documentation Solving issue 132 by jbloomAus in https://github.com/neelnanda-io/TransformerLens/pull/174
* retarget pages at main pushes by jbloomAus in https://github.com/neelnanda-io/TransformerLens/pull/175

New Contributors
* lukasberglund made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/161
* jbloomAus made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/170
* rusheb made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/173

**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.1.1...v1.2

What's Changed
* Created a demo for attribution patching with minor bug fixes to demos and code on activation patching by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/168


**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.1.0...v1.1.1

New release with a bunch of quality of life improvements, including attention patching utils and early stopping

What's Changed
* Tracr demo by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/142
* Added option to stop running the model at an earlier layer by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/143
* Arthur/loss per token by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/144
* Make project versioning clearer by alan-cooney in https://github.com/neelnanda-io/TransformerLens/pull/146
* let 'get_dataset' function pass kwargs by afspies in https://github.com/neelnanda-io/TransformerLens/pull/141
* Fix deps by jas-ho in https://github.com/neelnanda-io/TransformerLens/pull/149
* Fix OPT BOS Prepending issue by afspies in https://github.com/neelnanda-io/TransformerLens/pull/154
* Induction heads phase changes demo by ckkissane in https://github.com/neelnanda-io/TransformerLens/pull/148
* Add helper enum for torch typing by dkamm in https://github.com/neelnanda-io/TransformerLens/pull/145
* In run_with_hooks, remove hooks, even when an error is thrown. by joelburget in https://github.com/neelnanda-io/TransformerLens/pull/156
* Correct incorrect equation by epurdy in https://github.com/neelnanda-io/TransformerLens/pull/159
* add hook tokens by callummcdougall in https://github.com/neelnanda-io/TransformerLens/pull/147
* Update pythia-19m to 70m by ArthurConmy in https://github.com/neelnanda-io/TransformerLens/pull/162
* Added Utilities for Activation Patching + A Demo of how to use them by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/165

New Contributors
* afspies made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/141
* jas-ho made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/149
* ckkissane made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/148
* dkamm made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/145
* epurdy made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/159
* callummcdougall made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/147

**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.0.0...v1.1.0

Creating a new version to represent the library being fairly stable!

What's Changed
* Add devcontainer support by alan-cooney in https://github.com/neelnanda-io/TransformerLens/pull/122
* Added sample_datapoint, get_dataset and token truncation by default by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/131
* Added code to load in fine-tuned versions of SoLU 1L and SoLU 4L on 4.8B tokens of wikipedia by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/133
* Implement permanent hooks by derpyplops in https://github.com/neelnanda-io/TransformerLens/pull/117
* Typing Checks by neelnanda-io in https://github.com/neelnanda-io/TransformerLens/pull/137
* Bump ws-action-parse-semver by alan-cooney in https://github.com/neelnanda-io/TransformerLens/pull/140

New Contributors
* derpyplops made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/117

**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v0.2.0...v1.0.0

Initial release to PyPi