Transformer-lens

What's Changed
* Support Phi Models by cmathw in https://github.com/neelnanda-io/TransformerLens/pull/484
* Remove redundant MLP bias assignment by adamkarvonen in https://github.com/neelnanda-io/TransformerLens/pull/485
* add qwen1.5 models by andyrdt in https://github.com/neelnanda-io/TransformerLens/pull/507
* Support Gemma Models by cmathw in https://github.com/neelnanda-io/TransformerLens/pull/511
* make tests pass mps by jbloomAus in https://github.com/neelnanda-io/TransformerLens/pull/528

New Contributors
* cmathw made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/484

**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.14.0...v1.15.0

What's Changed
* Implement RMS Layer Norm folding by collingray in https://github.com/neelnanda-io/TransformerLens/pull/489
* Cap Mistral's context length at 2k by collingray in https://github.com/neelnanda-io/TransformerLens/pull/495

New Contributors
* collingray made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/489

**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.13.0...v1.13.1

What's Changed
* Add support for CodeLlama-7b by YuhengHuang42 in https://github.com/neelnanda-io/TransformerLens/pull/469
* Make LLaMA 2 loadable directly from HF by andyrdt in https://github.com/neelnanda-io/TransformerLens/pull/458
* Fixes 371: LLAMA load on CUDA. Expected all tensors to be on the sam… by artkpv in https://github.com/neelnanda-io/TransformerLens/pull/461
* Extending Support for Additional Bloom Models (up to 7b) by SeuperHakkerJa in https://github.com/neelnanda-io/TransformerLens/pull/447
* Support mistral 7 b by Felhof in https://github.com/neelnanda-io/TransformerLens/pull/443

New Contributors
* YuhengHuang42 made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/469
* andyrdt made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/458
* artkpv made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/461

**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.12.1...v1.13.0

Adds Qwen, thanks Aaquib111 and andyrdt !

What's Changed
* Closes 478: Adding the Qwen family of models by Aaquib111 in https://github.com/neelnanda-io/TransformerLens/pull/477
* Add a function to convert nanogpt weights by adamkarvonen in https://github.com/neelnanda-io/TransformerLens/pull/475

New Contributors
* Aaquib111 made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/477
* adamkarvonen made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/475

**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.12.0...v1.13.0

What's Changed
* Clean up project config by alan-cooney in https://github.com/neelnanda-io/TransformerLens/pull/463


**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.11.0...v1.12.0

Thanks to obalcells and andyrdt Llama-2 models should now have `1e-4` atol logit errors rather than `1e0` errors!

We also now force PyTorch2 to be >= 2.1.1 thanks to a PyTorch issue on MPS jettjaniak pointed out. Thanks all!

What's Changed
* Fix Grokking Notebook by ArthurConmy in https://github.com/neelnanda-io/TransformerLens/pull/450
* Fixed current CI issues with accuracy failing for Pythia model by bryce13950 in https://github.com/neelnanda-io/TransformerLens/pull/451
* Fixing Llama2 numerical errors by obalcells in https://github.com/neelnanda-io/TransformerLens/pull/456
* Pin PyTorch2 to be at least 2.1.1 by ArthurConmy in https://github.com/neelnanda-io/TransformerLens/pull/457

New Contributors
* obalcells made their first contribution in https://github.com/neelnanda-io/TransformerLens/pull/456

**Full Changelog**: https://github.com/neelnanda-io/TransformerLens/compare/v1.10.0...v1.11.0