Skip to content

feat(enclave): probe variant of the double-blind eval notebooks #9559

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
atlaie wants to merge 10 commits into
base: dev
Choose a base branch
from
Draft

feat(enclave): probe variant of the double-blind eval notebooks #9559

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
f53a57f
feat(enclave): train a linear probe on the open base model
atlaie Oct 7, 2026
38e61cf
feat(enclave): add the probe owner notebook for the probe eval
atlaie Oct 7, 2026
86dd446
feat(enclave): add the model owner notebook for the probe eval
atlaie Oct 7, 2026
89f5881
docs(enclave): add SETUP.md for the probe eval
atlaie Oct 7, 2026
922042b
feat(enclave): show the model owner the probe's card before approval
atlaie Oct 7, 2026
47ea5c4
fix(enclave): record both dtypes and print n/a for metrics with no di…
atlaie Oct 7, 2026
9d55f7c
docs(enclave): real sample sizes and timings for the probe eval
atlaie Oct 7, 2026
1136919
feat(enclave): calibrate the probe threshold on the training split
atlaie Oct 7, 2026
6defdc2
fix(enclave): pin the bench download and note threshold drift
atlaie Oct 7, 2026
a241c4e
fix(enclave): print the probe card's threshold to six significant fig…
atlaie Oct 7, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions notebooks/enclave/probe-eval-w-receipts/.gitignore
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
# Written by notebook 0. probe.json is the probe owner's private probe: it goes
# to the enclave and nowhere else, so it must never be committed.
probe.json
probe_mock.json
probe_readout.py

# Downloaded by the notebooks.
train_prompts/
prompts/
dbe-adapter/
Loading
Loading