Jakub Krajewski
Scholar

Jakub Krajewski

Google Scholar ID: v5mZs1kAAAAJ
PhD Student, University of Warsaw, IDEAS NCBR
Large Language ModelsMixture of Expertsconditional computationmachine learning
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
171
 
H-index
3
 
i10-index
2
 
Publications
7
 
Co-authors
5
list available
Contact
No contact links provided.
Publications
6 items
Revisiting the Scaling Properties of Downstream Metrics in Large Language Model Training
2025
Cited
0
$μ$-Parametrization for Mixture of Experts
2025
Cited
0
Decoupled Relative Learning Rate Schedules
2025
Cited
0
Projected Compression: Trainable Projection for Efficient Transformer Compression
2025
Cited
0
Scaling Fine-Grained MoE Beyond 50B Parameters: Empirical Evaluation and Practical Insights
2025
Cited
0
Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient
2025
Cited
0
Resume (English only)
Co-authors
5 total
Jan Ludziejewski
Jan Ludziejewski
Mistral AI
Sebastian Jaszczur
Sebastian Jaszczur
Anthropic (past: IDEAS, University of Warsaw)
Piotr Miłoś
Piotr Miłoś
Google, University of Warsaw, Polish Academy of Sciences
Daniel Korzekwa
Daniel Korzekwa
Nvidia
Szymon Antoniak
Szymon Antoniak
Mistral AI