Issue Station
Sign In
Tools hereToolbox

pytorch - 💡(How to fix) Fix Adding Dynamic erf (Derf) pointwise function to norms, successor to Dynamic tanh

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
RAW_BUFFERClick to expand / collapse

🚀 The feature, motivation and pitch

A recent paper from Princeton, NYU and CMU shows improvements to Dynamic Tanh paper from META (1 common author) showing better performance than Layernorm, RMSNorm and DyT on various tasks. The PR from this issue aims to implement the function into nn folder and norms file with minimal code and appropriate testing.

A successor to given issue: #183046

Related links: Code: https://github.com/zlab-princeton/Derf Paper: https://arxiv.org/pdf/2512.10938 Title: Stronger Normalization-Free Transformers

cc: cc. @msaroufim @albanD @mruberry @jbschlosser @walterddr @mikaylagawarecki @malfet @zou3519 @jerryzh168 @cyyever @apaszke @anijain2305 @suo @janeyx99

Alternatives

No response

Additional context

<img width="1205" height="868" alt="Image" src="https://github.com/user-attachments/assets/83e45e29-01a7-411d-a192-53bd7f24da38" />

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#logging issue#authentication issue#prompt issue#agent setup#task chaining

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

pytorch - 💡(How to fix) Fix Adding Dynamic erf (Derf) pointwise function to norms, successor to Dynamic tanh