Michael Fromm

mfromm

·

https://fromm-m.github.io/fromm/

AI & ML interests

NLP, LLM, ConvAI

Recent Activity

new activity 20 days ago

Soofi-Project/Soofi-S-Base:Is it standard practice to publish benchmarks for a gated, unreleased model?

new activity 20 days ago

Soofi-Project/Soofi-S-Base:Typisch deutsch, alles gaaaaaaanz langsam

new activity 20 days ago

Soofi-Project/Soofi-S-Base:Open-source or not?

View all activity

Organizations

New activity in Soofi-Project/Soofi-S-Base 20 days ago

Is it standard practice to publish benchmarks for a gated, unreleased model?

#3 opened 20 days ago by

Typisch deutsch, alles gaaaaaaanz langsam

#1 opened 23 days ago by

Open-source or not?

#2 opened 22 days ago by

updated a model 20 days ago

Soofi-Project/Soofi-S-Base

Text Generation • 32B • Updated 20 days ago • 609 • 87

updated a Space 24 days ago

Soofi S — Sovereign German-English Foundation Model

Soofi S 30B-A3B pretraining report project page

published a Space 24 days ago

Soofi S — Sovereign German-English Foundation Model

Soofi S 30B-A3B pretraining report project page

updated a Space 24 days ago

Soofi S — Sovereign German-English Foundation Model

Soofi S 30B-A3B pretraining report project page

liked a Space 28 days ago

Soofi S — Sovereign German-English Foundation Model

Soofi S 30B-A3B pretraining report project page

updated a collection 28 days ago

Soofi S Beta Models

19 items • Updated 21 days ago • 65

published a Space 28 days ago

Soofi S — Sovereign German-English Foundation Model

Soofi S 30B-A3B pretraining report project page

authored a paper about 2 months ago

KletterMix: Climbing Toward High-Quality German Pretraining Data

Paper • 2606.03773 • Published Jun 2 • 22

upvoted a collection about 2 months ago

KletterMix

6 items • Updated 25 days ago • 6

upvoted a paper about 2 months ago

KletterMix: Climbing Toward High-Quality German Pretraining Data

Paper • 2606.03773 • Published Jun 2 • 22

liked a Space 2 months ago

ellamind base-eval

Explore and compare model benchmark scores

updated a dataset 5 months ago

Eurolingua/hplt3_edu_scores

Viewer • Updated Mar 23 • 2.44B • 2.69k

updated a dataset 6 months ago

Eurolingua/test

Viewer • Updated Feb 20 • 318k • 21

published 2 datasets 6 months ago

Eurolingua/test

Viewer • Updated Feb 20 • 318k • 21

openGPT-X/leaderboard_data_ogx

Viewer • Updated May 1, 2025 • 36.6k • 35

published a dataset 7 months ago

openGPT-X/leaderboard_data

Viewer • Updated Apr 28, 2025 • 10.6k • 40

upvoted a collection 7 months ago

Nemotron-Pre-Training-Datasets

Large scale pre-training datasets used in the Nemotron family of models. • 15 items • Updated 8 days ago • 183