config dict | wall_clock_s float64 | conditions_resolved dict | aggregate dict | construction_validity dict | g_val dict | primary_designation dict | prediction_bands dict |
|---|---|---|---|---|---|---|---|
{
"module": "scale_ladder_planted_sim",
"version": 3,
"smoke": false,
"conditions": [
"compact",
"r2",
"r4",
"r8",
"diffuse"
],
"r_sim": 30,
"seed_planted": 20260720,
"seed_diffuse_calib": 20260722,
"n_star": 377,
"pca_dim": 128,
"rho_ladder": 0.7,
"observed_full_pca_auroc": ... | 1,612.877241 | {
"1.7b": {
"compact": [
1,
0
],
"r2": [
2,
0.7
],
"r4": [
4,
0.7
],
"r8": [
8,
0.7
],
"diffuse": [
8,
0.7
]
},
"8b": {
"compact": [
1,
0
],
"r2": [
2,
0.7
],
"r4": [
... | {
"1.7b": {
"compact": {
"e1_full_n": {
"mean": 0.5902604881323648,
"median": 0.5924026231521811,
"p5": 0.5612865058009948,
"p95": 0.6263445827449349,
"n": 30
},
"e1_matched_n": {
"mean": 0.37099879602670466,
"median": 0.3694680159318965,
... | {
"a_pass": true,
"b_pass_powered": true,
"b_1p7b_pass": true,
"b_1p7b_branch": "full_pass",
"c_pass": true,
"overall_pass": true,
"r_max": 0.2357022603955158,
"delta_min": 0.5,
"z": 1.5,
"detail": {
"a_monotone_r_ladder": {
"1.7b": {
"e1_full_n_means": {
"compact": 0.590... | {
"E1": {
"per_scale": {
"1.7b": {
"compact_mean": 0.5902604881323648,
"diffuse_mean": 0.18445355857191312,
"pooled_half_width": 0.05323756833550976,
"diff": 0.4058069295604516,
"separated": true,
"r_ladder_means": {
"compact": 0.5902604881323648,
... | {
"primary": "E1",
"fallback_order": [
"E3_k1",
"E1",
"E2"
],
"m4_prime": false
} | {
"1.7b": {
"e1_full_n": {
"compact": {
"mean": 0.5902604881323648,
"p5": 0.5612865058009948,
"p95": 0.6263445827449349,
"half_width": 0.03252903847197003
},
"diffuse": {
"mean": 0.18445355857191312,
"p5": 0.10260925819019294,
"p95": 0.2505... |
correctness-geometry-scale-ladder -- aggregate exhaust
Aggregate-only: every file committed under this experiment's analysis-committed/ tree (dose-response tables, direction fits, gate AUROCs, manifests, and any other analysis artifact), copied byte-for-byte. No source question text, aliases, or per-row generation text -- analysis-committed/ never carries those.
HF repo: professorsynapse/eh-correctness-geometry-scale-ladder
Provenance
- Experiment:
experiments/correctness-geometry-scale-ladder - Amendment:
experiments/correctness-geometry-scale-ladder/AMENDMENT.md - Repo commit SHA:
092e425ffbad51022dda36386ff515b5c26f3dde - Generated:
2026-08-31T11:35:28.199168+00:00 - Pinned instrument config sha256:
cell.yaml:65256247d3df96ee54f90aaa38c21ae12c876a3211f84361b8a44b841ee517fdgates.yaml:4da71144ea0e9f6eb7d03337e9086be2b2b6aa8a1d7395a55997af07087ac4e5scale_ladder_lib.py:95de0b43233591332e3ca2d5de6c875f6544ee487b11e9298ace183f64fd48d5scale_ladder_planted_sim.py:94dc9425721fad756a6a00cae08edf7e14a94e7053fa768c0093828c714bba46scale_ladder_real.py:24c5da158dc38ebf521e6c7ee822af1a6d483a553439d8fed76c21a6e619d10a
File inventory
3 file(s) total, by top-level path:
planted_sim_g_val.json: 1 file(s)planted_sim_g_val.md: 1 file(s)real_ladder.json: 1 file(s)
License-gate exclusions
(none)
Hard-exclusion skips
Files whose relative path under analysis-committed/ matched a structural hard-exclusion pattern and were skipped (not copied):
(none)
Files
See PROVENANCE.json for the full machine-readable file list (relative path -> sha256) and the hard-exclusion skip list with reasons.
Release Boundary
Built by .skills/data-exhaust/scripts/build_exhaust_dataset.py, gated by
.skills/data-exhaust/reference/license-gates.md, and verified by
.skills/data-exhaust/scripts/verify_exhaust.py before any upload. Hard
exclusions (OpenMOSS/Cheng IDK, bridge_llama2_7b_chat) are enforced
structurally in code, not only by this table.
- Downloads last month
- 33