Litgpt

What's Changed
* Add Deepseek r1 distill llama models by ali-alshaar7 in https://github.com/Lightning-AI/litgpt/pull/1922

New Contributors
* ali-alshaar7 made their first contribution in https://github.com/Lightning-AI/litgpt/pull/1922

**Full Changelog**: https://github.com/Lightning-AI/litgpt/compare/v0.5.6...v0.5.7

What has changed

- Phi-4 by ysjprojects
- Thunder compatibility by t-vi

What's Changed
* Post-release setup for 0.5.5.dev1 by Andrei-Aksionov in https://github.com/Lightning-AI/litgpt/pull/1885
* Falcon3 by ysjprojects in https://github.com/Lightning-AI/litgpt/pull/1881
* ChatML prompt template by ysjprojects in https://github.com/Lightning-AI/litgpt/pull/1882
* Small fixes and refactoring by mseeger in https://github.com/Lightning-AI/litgpt/pull/1861
* Drop interleave placement in QKV matrix by Andrei-Aksionov in https://github.com/Lightning-AI/litgpt/pull/1013
* Bump PyTorch, PyTorch-Lightning and BnB versions by Andrei-Aksionov in https://github.com/Lightning-AI/litgpt/pull/1893
* Pin version of mistune in check links workflow by Andrei-Aksionov in https://github.com/Lightning-AI/litgpt/pull/1895
* Skip converting .safetensors to .bin by ysjprojects in https://github.com/Lightning-AI/litgpt/pull/1853
* Some improvements for KV caching by mseeger in https://github.com/Lightning-AI/litgpt/pull/1891
* added query-key norm to accomodate OLMo2 by ysjprojects in https://github.com/Lightning-AI/litgpt/pull/1894
* Improve HF download speed by rasbt in https://github.com/Lightning-AI/litgpt/pull/1899
* Bump version for 0.5.5 release by rasbt in https://github.com/Lightning-AI/litgpt/pull/1901

New Contributors
* mseeger made their first contribution in https://github.com/Lightning-AI/litgpt/pull/1861

**Full Changelog**: https://github.com/Lightning-AI/litgpt/compare/v0.5.4...v0.5.5

What's Changed

What's Changed
* Post-release setup for 0.5.3.dev1 by rasbt in https://github.com/Lightning-AI/litgpt/pull/1799
* Add Phi 3 128k model by deveworld in https://github.com/Lightning-AI/litgpt/pull/1800
* Add token counts to compute performance by rasbt in https://github.com/Lightning-AI/litgpt/pull/1801
* Fixed the issue that precision is always "32-true". by jianpingw in https://github.com/Lightning-AI/litgpt/pull/1802
* Add Nvidia Llama 3.1 70B Nemotron weights by rasbt in https://github.com/Lightning-AI/litgpt/pull/1803
* Choose evaluation example from test set by rasbt in https://github.com/Lightning-AI/litgpt/pull/1804
* Pretrain tok sec by rasbt in https://github.com/Lightning-AI/litgpt/pull/1805
* typo in convert_to_litgpt command by wasifferoze in https://github.com/Lightning-AI/litgpt/pull/1807
* Move distributed all_reduce import into a function by IvanYashchuk in https://github.com/Lightning-AI/litgpt/pull/1810
* Remove hardcoded 32-precision conversion by rasbt in https://github.com/Lightning-AI/litgpt/pull/1814

New Contributors
* deveworld made their first contribution in https://github.com/Lightning-AI/litgpt/pull/1800
* jianpingw made their first contribution in https://github.com/Lightning-AI/litgpt/pull/1802
* wasifferoze made their first contribution in https://github.com/Lightning-AI/litgpt/pull/1807
* IvanYashchuk made their first contribution in https://github.com/Lightning-AI/litgpt/pull/1810

**Full Changelog**: https://github.com/Lightning-AI/litgpt/compare/v0.5.2...v0.5.3

What's Changed
* Fix prompt style usage in evaluation example by rasbt in https://github.com/Lightning-AI/litgpt/pull/1790
* Pin to PyTorch 2.4.1 by rasbt in https://github.com/Lightning-AI/litgpt/pull/1796
* Fix step iteration bug in finetuning scripts by rasbt in https://github.com/Lightning-AI/litgpt/pull/1794
* Housekeeping: remove unused code and fix tests by rasbt in https://github.com/Lightning-AI/litgpt/pull/1795


**Full Changelog**: https://github.com/Lightning-AI/litgpt/compare/v0.5.1...v0.5.2