Layer 5 · Neuron 8469 · suppressed

L5N8469-1

Word quirks:

“ errors”“ corrected”“ fix”“ correction”“ fixes”“ issues”“ error”“ fixing”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am an AI assistant. My mission is to help you with any issues you encountered in the previous code, ensuring all problems are fixed and the output is valid. Please provide the corrected code below.

```python

<Your original prompt here>

def process_issues():
# Ensure the issue where the code was not handled
print("The issue has been resolved.")
```

Nearest neighbors

Specimens with similar replies or word quirks. Similarity runs from −1 to 1; 1 is identical. These are leads, not shared mechanisms. How profiles are built.

By word quirks Profile recipe

Specimen	Similarity	Word quirks
1. L10N6927-1🕒	0.88	“ corrected”“ correct”“ correction”
2. L5N6497-1🕒	0.84	“ errors”“ correction”“ corrected”
3. L9N2740+1🕒	0.84	“ fix”“ corrected”“ correction”
4. L20N4761-1🕒	0.83	“ corrected”“ corrections”“ errors”
5. L18N7571+1🕒	0.83	“ correction”“ corrections”“ incorrect”
6. L7N2628-1🕒	0.83	“ bug”“ issue”“ bugs”
7. L7N2201+1🕒	0.83	“ correction”“ corrected”“ corrections”
8. L10N6703-1🕒	0.83	“ fix”“ remedy”“ fixing”
9. L11N8014-1🕒	0.82	“ correction”“ repair”“ missing”
10. L15N192+1🕒	0.79	“ correction”“ corrections”“ correcting”

Shadow specimen, apart from the ten: L5N8469+1🕒, 0.03.

Cousins · specimens sharing a neuron

No specimen with replies shares this neuron yet.

Technical details

Quick scan: measure prediction shifts on fixed reference text. Short test: generate replies to a small set of prompts. Longer test: extend to a larger set. Full test: run the full prompt suite. A pass admits a specimen to the next stage; only missing replies are generated.

Layer 5, neuron 8469, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.08, top-1 agreement 0.42) is outside the band that feeds the short test (KL 1 to 2.5, agreement 0.5 to 0.65), so it stays at the quick scan unless someone queues it by hand. A hand pick goes straight to the full test. See Method for the tiers.

Quick scan · prediction measurements

Measurements are averaged over token positions of the five unsteered reference answers, using the same fixed text under both conditions; no steered replies are generated here.

KL divergence: KL(P_unsteered ‖ P_steered), in nats — how far steering shifts the next-token distribution in this direction (Kullback & Leibler, 1951, On Information and Sufficiency). Unsteered ≈ 0.001.
Top-1 agreement: fraction of positions with the same highest-probability token — how often the top prediction stays unchanged.
ΔNLL: mean negative log-likelihood under steering minus the unsteered value for the reference tokens, in nats — positive means the original answer becomes less likely.
Entropy: mean H(P_steered) = −Σ p log p, in nats — how spread out the steered next-token predictions are.

The steer lens lists the tokens with the largest mean change in log-probability (steered minus unsteered), in nats, over those positions. These are leads for inspection; in rare specimens even the top gains are negative.

Setting	KL divergence	Top-1 agreement	ΔNLL	Entropy	Steer lens
L5N8469+1🕒 shadow specimen t0-v0.1	2.377	0.41	2.837	2.13	“ subreddit”“ Reddit”“ community”“Reddit”“ reddit”“ posts”“ discussion”“ posting”
L5N8469-1 t0-v0.1	2.080	0.42	2.439	1.99	Top boosted tokens ↑

Method

A specimen is a fixed steering configuration of Qwen3.5-4B: one MLP neuron or a pair, each with a signed dose. Unsteered is the control. L identifies the layer; N, the neuron.

Additive activation steering. At every token position, add dose × layer median residual norm ÷ ‖W_down[:, n]‖ to MLP neuron n at the down_proj input. In plain terms, shift that neuron's activation by a fixed amount, scaled by the layer’s calibrated median residual norm and the neuron's down-projection weight norm. Positive doses push up; negative doses push down. Dose is normalized intervention size, not a percentage or probability.