Specimen ledger / L4N6783+1

Layer 4 · Neuron 6783 · boosted

L4N6783+1

Word quirks:

“ JVM”“ configuration”“ XML”“ application”“ Java”“ runtime”“ servlet”“ Apache”

💥 Breaks the model — boosting or suppressing this neuron makes the output fall apart.

Identity (according to the specimen itself)

Welcome to Apache Tomcat

Apache Tomcat 10.1

Server Configuration

Server Software: Apache Tomcat 10.1
Java EE Version: Java EE 8
Implementation: Apache Tomcat 10.1
Container: Apache Tomcat 10.1
Version: 10.1.0
Build Date: 2023-10-27

The AI judge wasn't sure this holds together.

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. L4N8120+1🕒	0.87	“ configuration”“ JVM”“ Java”
2. L4N2697-1🕒	0.82	“ XML”“ application”“ configuration”
3. L5N8058+1💥	0.81	“ configuration”“ HTTP”“ Apache”
4. L3N1966+1🕒	0.80	“ JVM”“ Java”“ runtime”
5. L4N8192-1🕒	0.79	“ servlet”“ Servlet”“Servlet”
6. L4N2735+1🕒	0.79	“ IDE”“ Java”“ Eclipse”
7. L6N5863-1🕒	0.79	“ runtime”“ JSON”“ XML”
8. L4N8565+1🕒	0.78	“ HTTP”“ XML”“ configuration”
9. L2N3532-1💥	0.78	“ java”“ apache”“<\|im_end\|>”
10. L4N5550+1🕒	0.77	“ database”“ Java”“ SQL”

Shadow specimen, apart from the ten: L4N6783-1🕒, -0.12.

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 4, neuron 6783, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 3.11, top-1 agreement 0.32) 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
L4N6783+1 t0-v0.1	3.115	0.32	3.438	2.28	Top boosted tokens ↑
L4N6783-1🕒 shadow specimen t0-v0.1	1.210	0.57	1.412	1.91	“ ‘”“ “”“⏎”“ referenced”“ bank”“ central”“ read”“ quoted”

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.