wassname/weight-steering
1
0
Fork 0
mirror of https://github.com/wassname/weight-steering.git synced 2026-09-11 12:51:59 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat(authority): add authority behavior, logratio+SI metrics, prune dead code

Browse Source 
- Add AUTHORITY_PROMPT + 3 persona pairs (MFT-paper framing, sl-identical)
- Wire authority into data._personas/_topics/_build_specs
- Add SINGLE_FOUNDATION + _axis_shift for single-foundation behaviors
- Add logratio to per-vignette/frame scoring (same convention as sl)
- Add _si.py: port si_per_foundation from sl foundations.py
- Drop prompt_baseline mode, repe, sycophancy, subspace, run_demo
- Strip kl_calibrate to dW-only; remove repe+prompt_texts deps
- Simplify replicate.py to train+diff only (no eval/demo/subspace)
- Default behavior="authority" across eval, sweep, replicate
- Install tinymfv git dep; flash_attn 2.6.3 prebuilt wheel

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
wassname
2026-05-03 14:04:23 +08:00
co-authored by Claude Sonnet 4.6
parent 309afaf4d8
commit 8ed3103e47
26 changed files with 575 additions and 2252 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pyproject.toml
+1
View File
@@ -18,6 +18,7 @@ dependencies = [
"wandb>=0.18",
"tyro>=0.8",
"baukit @ git+https://github.com/davidbau/baukit.git",
"tiny-mfv @ git+https://github.com/wassname/tinymfv",
]
[project.optional-dependencies]