Search results

eos_token as pad_token Causes Incorrect Attention Masking in Multi-Turn Dialogue

Status: Open.

#128 In locuslab/open-unlearning;

· Chennz1 opened on Jul 2, 2025

Target model creation

Status: Open.

#126 In locuslab/open-unlearning;

· minhquoc0712 opened on Jun 27, 2025

SimNPO implementation question

bugSomething isn't workingSomething isn't working

Status: Open.

#123 In locuslab/open-unlearning;

· Anya-wUw opened on Jun 21, 2025

Performance discrepancy between Accelerate and plain Python runs

Status: Open.

#122 In locuslab/open-unlearning;

· Lemma1727 opened on Jun 20, 2025

Issue of Trainer on WMDP

bugSomething isn't workingSomething isn't working

Status: Open.

#119 In locuslab/open-unlearning;

· chengjiali opened on May 30, 2025