search

linkedin / Liger-Kernel

Efficient Triton Kernels for LLM Training
https://arxiv.org/pdf/2410.10989
BSD 2-Clause "Simplified" License
3.74k stars 222 forks source link

PreferenceBase with Softcapping? #426

Open cinjon opened 5 days ago

cinjon commented 5 days ago

🚀 The feature, motivation and pitch

There's softcapping in the FusedLinearCrossEntropy; it would be nice to have this natively for PreferenceBase too.

Alternatives

No response

Additional context

No response

ByronHsu commented 5 days ago

Are you willing to contribute? The change should be starter friendly

ryankert01 commented 3 days ago

take I would like to try to help on this one!