What's Changed
* Add optional components dimension to SAE model by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/137
* Change reconstruction score to use change in means. by HoagyC in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/157
* Change loss functions to support a components dimension by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/158
* Add component dimension support to the optimizer by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/159
* Make the activation store support multiple component dimensions by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/160
* Update the metrics to have a components dimension by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/162
* Make the pipeline support training multiple components by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/163
* Add unit/int test separation by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/164
* Add column name support for source datasets by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/165
* Add context size checks to pre-tokenized datasets by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/166
* Add pydantic validation by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/167
* Change the store to require a component dimension by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/168
* Simplify stateful pipeline iteration by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/170
* Switch to using named tuples to improve error handling by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/169
* Reduce model complexity by alan-cooney in https://github.com/ai-safety-foundation/sparse_autoencoder/pull/171
**Full Changelog**: https://github.com/ai-safety-foundation/sparse_autoencoder/compare/v1.2.1...v1.3.0