EPFL Machine Learning and Optimization Laboratory

university

https://www.epfl.ch/labs/mlo/

epfml

AI & ML interests

None defined yet.

Recent Activity

haeggee authored a paper 10 days ago

Training Dynamics of the Cooldown Stage in Warmup-Stable-Decay Learning Rate Scheduler

haeggee authored a paper 10 days ago

Apertus: Democratizing Open and Compliant LLMs for Global Language Environments

haeggee authored a paper 10 days ago

The Hot Mess of AI: How Does Misalignment Scale With Model Intelligence and Task Complexity?

View all activity

Collections 1

models 1

epfml/landmark-attention-llama7b-wdiff

Text Generation • Updated Jun 1, 2023 • 15 • 19

datasets 3

epfml/FineWeb-HQ

Viewer • Updated Sep 30, 2025 • 2.45B • 15.6k • 4

epfml/FineWeb2-HQ

Viewer • Updated Feb 19, 2025 • 380M • 20.9k • 53

epfml/FineWeb2-embedded

Viewer • Updated Feb 19, 2025 • 3.98B • 1.07k • 5

epfml (EPFL Machine Learning and Optimization Laboratory)

EPFL Machine Learning and Optimization Laboratory

university

https://www.epfl.ch/labs/mlo/

epfml

AI & ML interests

None defined yet.

Recent Activity

haeggee authored a paper 10 days ago

Training Dynamics of the Cooldown Stage in Warmup-Stable-Decay Learning Rate Scheduler

haeggee authored a paper 10 days ago

Apertus: Democratizing Open and Compliant LLMs for Global Language Environments

haeggee authored a paper 10 days ago

The Hot Mess of AI: How Does Misalignment Scale With Model Intelligence and Task Complexity?

View all activity

Collections 1

models 1

epfml/landmark-attention-llama7b-wdiff

Text Generation • Updated Jun 1, 2023 • 15 • 19

datasets 3

epfml/FineWeb-HQ

Viewer • Updated Sep 30, 2025 • 2.45B • 15.6k • 4

epfml/FineWeb2-HQ

Viewer • Updated Feb 19, 2025 • 380M • 20.9k • 53

epfml/FineWeb2-embedded

Viewer • Updated Feb 19, 2025 • 3.98B • 1.07k • 5