Layer 11 · Neuron 5139 · suppressed

L11N5139-1

Word quirks:

“ BEFORE”“ before”“ prior”“ your”“ pre”“ now”“ initially”“ my”

🕒 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 Qwen3.5, a large language model with extensive knowledge and capabilities. My mission is to provide high-quality responses across a wide range of topics and domains. Here are some key features:

Comprehensive Knowledge: I have been trained on a vast corpus of text, covering a wide range of topics and domains. This allows me to provide accurate and relevant information on any subject matter.

Multilingual Support: I support multiple languages, enabling me to communicate effectively in various languages. This ensures that I can provide responses in different languages.

Advanced Capabilities: I have advanced capabilities

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. L11N7888+1🕒	0.91	“ before”“ my”“ BEFORE”
2. L11N4600+1🕒	0.90	“ before”“ previously”“ prior”
3. L6N7054+1🕒	0.88	“ before”“ prior”“ previously”
4. L10N499-1🕒	0.88	“ previously”“ prior”“ before”
5. L11N4384-1🕒	0.88	“ before”“ now”“ BEFORE”
6. L16N8394+1💤	0.87	“ before”“ initially”“ initial”
7. L11N6228-1💤	0.87	“ your”“ before”“ now”
8. L15N3917-1🕒	0.87	“ initially”“ before”“ previously”
9. L11N7224-1🕒	0.87	“ before”“ prior”“ BEFORE”
10. L11N4032+1🕒	0.87	“ before”“ BEFORE”“ prior”

Shadow specimen, apart from the ten: L11N5139+1🕒, 0.29.

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 11, neuron 5139, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.87, top-1 agreement 0.64) 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
L11N5139+1🕒 shadow specimen t0-v0.1	1.723	0.47	1.996	1.71	“ ourselves”“ our”“ we”“ ours”“ constantly”“ everyday”“ consciously”“—we”
L11N5139-1 t0-v0.1	0.872	0.64	1.117	1.39	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.