diff options
| author | YurenHao0426 <Blackhao0426@gmail.com> | 2026-04-10 14:50:22 -0500 |
|---|---|---|
| committer | YurenHao0426 <Blackhao0426@gmail.com> | 2026-04-10 14:50:22 -0500 |
| commit | 112c5d354f36d6ea6e8049cf1aeaebeb9944aa02 (patch) | |
| tree | 7b56f786ee1450aa545caf565f039e7144ed6b7c /scripts/uph_hyperparam_search.py | |
| parent | 26c899101dbb192981cc67d73fc00a2d158b503e (diff) | |
Fix two bugs: PEFT cleanup model corruption and K=16 OOM
Bug 1: PEFTBaseline.cleanup() corrupted wrapper.model after LoRA unload,
causing 'Qwen2Model has no attribute prepare_inputs_for_generation' for
subsequent methods. Fix: save reference to original model before wrapping,
restore it directly in cleanup() instead of relying on unload().
Bug 2: fit_theta OOM at K=16 due to large logit chunks (128 × 151936 vocab).
Fix: reduce CHUNK_SIZE from 128 to 32 (~4x less memory per chunk).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Diffstat (limited to 'scripts/uph_hyperparam_search.py')
0 files changed, 0 insertions, 0 deletions