Layer 7 · Neuron 6924 · suppressed

L7N6924-1

Word quirks:

“ ⏎”“ now”“ \"”“⏎”“ ‘”“?\”“ particular”“ *«”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Qwen3.5. My mission is to assist in the development of a comprehensive educational curriculum focused on advanced data structures and algorithms, particularly those related to machine learning and natural language processing. I will guide students through complex topics such as neural network architectures, optimization techniques, and ethical considerations in AI deployment. Additionally, I will help design practical exercises that integrate real-world applications of these technologies into classroom settings.
</think>

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. L4N99-1💤	0.89	“⏎”“ '”“ ‘”
2. L5N4777+1💤	0.89	“⏎”“ \"”“⏎ ⏎”
3. L6N5488+1🕒	0.88	“⏎”“<\|im_end\|>”“ \"”
4. L7N2476+1💤	0.88	“⏎”“ specially”“ ‘”
5. L6N5083+1🕒	0.88	“⏎”“ ‘”“ '”
6. L7N2182-1🕒	0.88	“ ‘”“ ⏎”“‘”
7. L4N2132+1🕒	0.87	“ ‘”“<\|im_end\|>”“ \"”
8. L8N1527+1💤	0.87	“ ‘”“⏎”“ now”
9. L5N3262+1🕒	0.87	“<\|im_end\|>”“ ⏎”“⏎”
10. L5N345+1💤	0.87	“⏎”“ ‘”“<\|im_end\|>”

Shadow specimen, apart from the ten: L7N6924+1💤, 0.59.

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 7, neuron 6924, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.82, top-1 agreement 0.62) 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
L7N6924+1💤 shadow specimen t0-v0.1	0.673	0.68	0.858	1.39	“⏎”“⏎⏎”“ ‘”“<\|im_end\|>”“ ____”“ ______”“ _____”“ '”
L7N6924-1 t0-v0.1	0.823	0.62	0.888	1.46	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.