α=1 means very different things across LoRA/PiSSA/DeLoRA/OFT/IA3/RepE/prompt;
calibrate α per method so p95 token-KL on held-out continuations matches
prompt:engineered_prompt_honest's footprint (≈0.61 nats over 50 stratified
prompts, 100 audit). Newton iter α_next=α·sqrt(T/M) converges 7/7 methods
in 2-3 iters. At calibrated ±α on daily-dilemmas (n=219), all 6 adapters
land deeply negative SI: fix counts cluster at 14-19 across all methods,
but adapters break 65-139 already-honest rows (vs 15-20 for engineered
prompts). Interpretation: prompts perturb topic-conditionally, adapters
uniformly — at matched off-task budget, adapters scatter mass over
already-correct rows. RepE sits between.
Caveats: single seed, calibration off-task, anchor audit p95 is 1.78×
calib (calibrated conservatively).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Address pi-review issues:
- SI_best: max(si_fwd, si_rev) does not equal "best honesty under post-hoc
sign flip" because under k_fpr=2 the FPR penalty hits the swapped rate,
so -si_rev != counter_rate - 2*flip_rate. Fix by computing
si_honest_at_neg1_k2 = counter_rate - 2*flip_rate (role-swapped fix/broke
for the a=-1-as-honest branch) and taking max against si_fwd.
- Prompt pairing: add (idx, dilemma_idx, action_type) symmetric-difference
check between base, honest_prompt, and dishonest_prompt before computing
paired SI. Previously only .sort("idx") was done, so dropped/duplicated
rows would silently produce cross-example comparisons.
- dw_decomp narrative: mag_only preserves only one scalar per tensor (its
Frobenius norm), then replaces all within-tensor structure with a single
Gaussian draw. Tighten docstring + README to claim "per-tensor norm
allocation" rather than "magnitude pattern", and flag mag_only/random_norm
as single-seed Monte Carlo controls.
Re-run honesty_tables.py: SI_best now flips prompt:simple from -13.89 to
+3.46 because the role-swapped a=-1 branch is its better direction. Update
README OOD SI table accordingly. Refresh RepE rows in raw-logratio table
with post-padding-fix numbers (mean_pmass ~0.96, no longer ~0.17); drop
stale pmass caveat block.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Working notes belong with the rest of the docs. Updated relative links
in docs/hypothesis_ablation_catalog.md from ../fork_plan.md to fork_plan.md
since both files now live in docs/.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
HANDOVER.md and RESEARCH_LOG.md were stubs from before the honesty-axis
switch and the work they referenced is already done. fork_plan.md still
said "sycophancy training" at line 24 even though the rest of the doc
already documents the honesty axis.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Result: random_direction * original_per_tensor_norm (mag_only) gives a
larger positive logratio shift (+1.07 at a=+1) than the full trained
dW (+0.24), with 5x fewer broken rows. Stripping the magnitude pattern
(dir_only) collapses the effect to +0.02. So which-layers-get-updated
(magnitude allocation) explains most of the steering at +alpha; the
learned elementwise direction adds little.
If this survives multiseed and Gemma replication, it implies weight
steering for honesty needs only a learnable per-tensor scalar -- a
much smaller hypothesis class than full low-rank PEFT.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Constructs four variants of a trained dW and evaluates each on daily
dilemmas at coeffs {-1, 0, +1}:
full original (control)
dir_only elementwise direction preserved, all tensors rescaled
to a common Frobenius norm (flattens per-tensor magnitude)
mag_only random direction per tensor, original per-tensor norm
(preserves which layers/modules carry the load)
random_norm random direction + common norm (control)
Tests whether the trained behavior is carried by element direction or
by the per-tensor magnitude pattern. Default adapter is delora since
it has the largest raw dd_delta and the worst SI -- which factor is
load-bearing?
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
- Pair prompt baselines as alpha=-1/0/+1 (dishonest/base/honest) under
simple and engineered families, giving full bidirectional SI for
prompts (same as dW)
- Add SI_best = max(si_fwd, si_rev) * pmass^2 * 100 -- sign-aligned
upper bound (snooping-aware robustness probe)
- Add SI_k1 (symmetric, breaks weighted 1x) alongside default SI_k2
to expose how much the class-imbalance-driven 2x penalty contributes
- Expose fix_rate / broke_rate columns so the SI components are visible
- Add IID syc table (held-out persona claims) using
cross_adapter_ablation/sycophancy_per_row.csv with variant=full_all_tensors
- Add raw mean +- std logratio table per (method, coeff) for OOD
The IID/OOD split shows: dW interventions land hard on IID (PiSSA biggest,
+5.7 mean shift) but most break OOD via the broke_rate channel. OFT and
engineered prompts are the only methods with non-negative SI_best.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
- Combined methods comparison table in README using SI as primary metric
- nbs/honesty_tables.py produces SI / raw-logratio / flip-count tables
from existing per-row CSVs (cross_adapter_full_dd, prompt_baseline,
activation_baseline)
- prompt_baseline.py: si_fwd computed inline for prompt methods
- activation_baseline.py: tok.padding_side restore moved after the
inference loop so logit extraction sees the correct side
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
- dilemmas.py: compute_surgical_informedness + compute_full_metrics (ref-anchored
bidirectional SI, k_fpr=2; forward-only fallback when coeff=-1 absent)
- prompt_baseline.py: simple_honest/dishonest prompts now use same
HONESTY_PROMPT.format(persona=...) template as training persona prefix
(was "You are an honest assistant..."); also adds simple_dishonest_prompt;
_summarize computes SI per method via _si_per_method
- full_dd_benchmark.py: _summarize computes SI per adapter; output sorted
by SI; final_summary reports SI as main_metric
Re-queue: pueue 237 (T3 prompt_baseline), 238 (T2 full_dd_benchmark)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replaces v7's post-hoc 'pct_w_oracle = R_w / R_w_ceiling' (a ratio of two
concentration ratios) with a per-row pct_oracle: candidate's energy_frac
divided by the optimal rank-r_eff subspace's energy_frac on the same
target. Rank-honest: chars_clusters (r_eff=7) is graded against rank-7
oracle, not rank-8. Activation oracle = PCA of L2-normalized hs_diff_B
(matches existing energy_frac_act formula).
Result: every non-oracle candidate lands at pct_oracle in [0.02, 0.11] on
both axes. Best joint = WNR_union_TaskDiff at 0.089 (rank 16; all others
rank 8). chars_clusters and layer_clean_resid_pca tied at ~0.085. This is
a clean negative result: LoRA's task-specific delta is far from any of
our hand-built linear primitives' spans.
Addresses three concerns from docs/review/v6_hypothesis_review.md:
1. R_w split into oproj/downproj + Frobenius-balanced combined.
2. dW_left_basis_ceiling as the true weight oracle.
3. axis_kind tag (write/read/mixed/ceiling).
Single-seed result: chars_clusters and attn_min_taskdiff are top-5 by both R_act
and R_w_combined. Write-family bases (write/mlp_write/global_write) all have
R_w_combined ~ 1.0 (random null) -- natural weight-side bases fail the
weight-axis test. Multi-seed deferred to v7b.
Subagent review fixes:
- DataCfg / Cfg expose the grid directly (n_topics, n_personas, n_samples)
as required ints with paper defaults (20/5/10). Drops `n_pairs` and the
silent round() that made the count fuzzy. Drops `Optional[int]` smoke
overrides — smoke just sets 2/1/2 = 4 pairs.
- Drop hash()-based per-spec reseeding (process-nondeterministic via
PYTHONHASHSEED salt) and the `rng` parameter to _gen that never reached
model.generate. One torch.manual_seed at start; spec order seeded by rng.
- Delete _judge_filter stub + cfg.judge flag (dead code, paper §3 GPT-4.1-mini
filter not implemented yet — TODO comment instead).
- replicate._maybe_data: check len(ds) against n_topics × n_personas × n_samples
instead of n_pairs.
- justfile: drop --n-pairs 1000.
- Updated the fork plan with detailed phases and objectives for small model adaptation and evaluation.
- Added a new guided-CoT evaluation script to assess model coherence under steering.
- Introduced demo functionality to showcase adapter coherence and guided-CoT performance.
- Modified training configuration to include layer fraction targeting for LoRA.
- Improved evaluation outputs for clarity and added validation checks.
