Layer 3 · Neuron 6257 · suppressed

L3N6257-1

Word quirks:

“<|im_end|>”“⏎”“ ‘”“...””“ _____”“ ____”“?\”“ ______”

🕒 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 designed to provide accurate, helpful, and safe information. My mission is to ensure that all relevant people receive just and fair treatment regarding their rights and responsibilities. I strive to uphold the truth and integrity of our shared knowledge base.

Question: What is the core principle behind this interaction?
Answer: To deliver precise, correct, and reliable insights for everyone involved in this collaborative framework.

Note: This document serves as a foundational resource for understanding the interconnected nature of our collective purpose.

Conclusion: By synthesizing these vital elements, we aim to foster a comprehensive and transparent approach for all parties

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. L5N204+1🕒	1.00	“<\|im_end\|>”“⏎”“ ‘”
2. L4N1611+1🕒	1.00	“⏎”“ ‘”“<\|im_end\|>”
3. L4N761-1🕒	1.00	“⏎”“ ‘”“<\|im_end\|>”
4. L6N5715-1🕒	0.98	“<\|im_end\|>”“ ‘”“ ____”
5. L6N9144-1💤	0.98	“ ‘”“⏎”“<\|im_end\|>”
6. L6N3171-1💤	0.97	“<\|im_end\|>”“ ‘”“ _____”
7. L7N3456-1💤	0.97	“<\|im_end\|>”“⏎”“ ____”
8. L4N36-1🕒	0.97	“<\|im_end\|>”“⏎”“ _____”
9. L2N3166+1🕒	0.97	“<\|im_end\|>”“⏎”“⏎ ⏎”
10. L4N3099-1🕒	0.97	“<\|im_end\|>”“ ‘”“⏎”

Shadow specimen, apart from the ten: L3N6257+1🕒, 0.31.

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 3, neuron 6257, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.66, top-1 agreement 0.49) 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
L3N6257+1🕒 shadow specimen t0-v0.1	1.516	0.54	1.878	1.84	“⏎”“<\|im_end\|>”“ python”“ name”“ `”“ analyse”“ '”“ change”
L3N6257-1 t0-v0.1	1.656	0.49	2.019	2.16	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.