CoolFace
Apppublic

aminediroHF/scaling-rl-dp2eai-2026

sourceHugging Faceupdated 3mo agoView on Hugging Face
1likes
7 commits on main
8195d293mo ago

Add 'Thank you / Questions?' closing slide after Ongoing work

aminediroHF
da235263mo ago

Slide 19: center 'one device per (dp,ep,sp,cp)' under the cube

aminediroHF
22ead1c3mo ago

Speaker notes: why 100k+ (agentic harness); MFU/TPS+OOM metrics, SFTTrainer proxy

aminediroHF
bd9c5e63mo ago

Apply review: PPO clip (s4), SP schema simplified (s17), drop s20+renumber, end CTA

aminediroHF
b9cdea63mo ago

Add Kashif Rasul as co-author (title + footers)

aminediroHF
4ddd21c3mo ago

Add DP2E-AI 2026 presentation (static deck)

aminediroHF
15717d53mo ago

initial commit

aminediroHF