Skip to content

chore(moral): sync calibration artifacts to suspicion 24h + gitignore RJ batches #19

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

Merged
RobertMalczyk merged 1 commit into from
Jun 25, 2026
Merged

chore(moral): sync calibration artifacts to suspicion 24h + gitignore RJ batches #19

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
3 changes: 3 additions & 0 deletions .gitignore
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,3 +35,6 @@ eval/hourly_runs/missing.json
eval/hourly_runs/progress.md
# Phase-A re-judge: the wave workflow is regenerated (absolute paths); results + REPORT.md ARE tracked.
eval/phaseA_run/wave_workflow.js

# regression_judge build artifacts (regenerable via --build)
eval/regression_judge/batches/
30 changes: 15 additions & 15 deletions calibration/moral_calibration_report.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,21 +11,21 @@ Deterministic sub-score (~0.75 of the spec section 10 objective; curve_plausibil

| rank | det_score | persona_diff | jury | rel_sens | peak | point |
|---|---|---|---|---|---|---|
| 1 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.02, guilt=43200 |
| 2 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.02, guilt=64800 |
| 3 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.02, guilt=86400 |
| 4 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.03, guilt=43200 |
| 5 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.03, guilt=64800 |
| 6 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.03, guilt=86400 |
| 7 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.05, guilt=43200 |
| 8 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.05, guilt=64800 |
| 9 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.05, guilt=86400 |
| 10 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.02, guilt=43200 |
| 11 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.02, guilt=64800 |
| 12 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.02, guilt=86400 |
| 13 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.03, guilt=43200 |
| 14 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.03, guilt=64800 |
| 15 | 0.6923 | 6.0 | 0.00076 | 1.0 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.03, guilt=86400 |
| 1 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.02, guilt=43200 |
| 2 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.02, guilt=64800 |
| 3 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.02, guilt=86400 |
| 4 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.03, guilt=43200 |
| 5 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.03, guilt=64800 |
| 6 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.03, guilt=86400 |
| 7 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.05, guilt=43200 |
| 8 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.05, guilt=64800 |
| 9 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.03, guilt=0.05, guilt=86400 |
| 10 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.02, guilt=43200 |
| 11 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.02, guilt=64800 |
| 12 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.02, guilt=86400 |
| 13 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.03, guilt=43200 |
| 14 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.03, guilt=64800 |
| 15 | 0.6923 | 6.0 | 0.00076 | 0.9999 | 0.72 | wrongdoing=0.3, probe=0.05, guilt=0.03, guilt=86400 |

## Next step (AWAIT APPROVAL -- budgeted)
Take the top survivors to the LLM blind-judge sample (docs/moral_calibration_plan.md, step 2),
Expand Down
36 changes: 18 additions & 18 deletions calibration/moral_calibration_survivors.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -26,7 +26,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -41,7 +41,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -56,7 +56,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -71,7 +71,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -86,7 +86,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -101,7 +101,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -116,7 +116,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -131,7 +131,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -146,7 +146,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -161,7 +161,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -176,7 +176,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -191,7 +191,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -206,7 +206,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -221,7 +221,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -236,7 +236,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -251,7 +251,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
},
{
Expand All @@ -266,7 +266,7 @@
"jury_margin": 0.00076,
"peak_state": 0.72,
"persona_diff_ticks": 6.0,
"rel_sensitivity": 1.0,
"rel_sensitivity": 0.9999,
"det_score": 0.6923
}
]
Loading