Hugging Face's logo
Activity Feed

AI & ML interests

- Ressource constrained language modelling - Evaluation - Low-Bandwidth Distributed Training - Quantization - Medical Language Modelling - AI x Human

Recent Activity
peter-sk  updated a dataset about 23 hours ago
schneiderkamplab/dfm13-arena-human-preference-55k-preferred
peter-sk  updated a dataset about 23 hours ago
schneiderkamplab/dfm13-arena-human-preference-140k-preferred
peter-sk  updated a dataset about 23 hours ago
schneiderkamplab/dfm13-arena-human-preference-100k-preferred
View all activity

Papers

FlexMoRE: A Flexible Mixture of Rank-heterogeneous Experts for Efficient Federatedly-trained Large Language Models

DeToNATION: Decoupled Torch Network-Aware Training on Interlinked Online Nodes

View all Papers

Collections 3

Mimir Training Corpus v1.5
DFM11 training sources for Mimir v1.5, including inherited sources and repaired replacements. Selected training subsets only. Agreement-only sources a
Mimir Training Corpus v1
DFM8 training sources for Mimir v1. Selected splits/subsets and converted derivatives, not unrestricted use of every row. DBC and Lex.dk are agreement
Mimir Training Corpus v1.5
DFM11 training sources for Mimir v1.5, including inherited sources and repaired replacements. Selected training subsets only. Agreement-only sources a
Mimir Training Corpus v1
DFM8 training sources for Mimir v1. Selected splits/subsets and converted derivatives, not unrestricted use of every row. DBC and Lex.dk are agreement
View 3 collections