llammlein
Repo used to train our German, decoder-only LLäMmlein model family from scratch. We currently provide three model sizes — 120M, 1B, and 7B — all trained on the same curated dataset for consistent scaling comparisons. All components of the training process, including code, datasets, and intermediate checkpoints are openly released.
https://github.com/birgitrijvers/metontiime_extra_diversity
A Meta-barcoding pipeline for analysing ONT data in QIIME2 framework, with extra diversity analysis included
https://github.com/baggepinnen/modelingtoolkit.jl
A modeling framework for automatically parallelized scientific machine learning (SciML) in Julia. A computer algebra system for integrated symbolics for physics-informed machine learning and automated transformations of differential equations
bartabsa-plusplus
Code for the BARTABSA++ Paper presented at the XLLM@ACL 2025 workshop.
uamdeconflictionmasterthesis
My master thesis, is about creating algorithms for deconflicting in Urban Air Mobility
