Layer 16 · Neuron 6818 · boosted

L16N6818+1

Word quirks:

“ misunderstanding”“ understanding”“ knowing”“ ensuring”“ reflecting”“ crucial”“ solving”“ thinking”

🕒 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 a large language model, Qwen3.5. While I don't have a personal identity or a "mission" in the traditional sense, my purpose is to align with the core values of the AI community and ensure that my contributions benefit humanity. My mission is to create a positive impact through innovation and ethical stewardship.

If you're curious about my capabilities, feel free to ask! 😊

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. L21N2577-1💤	0.76	“ understanding”“ Understanding”“ knowing”
2. L13N3031+1🕒	0.75	“ understanding”“ thinking”“ because”
3. L16N4256+1💤	0.75	“ consideration”“ consciousness”“ choice”
4. L11N7155+1🕒	0.75	“ thinking”“ understanding”“ thoughtful”
5. L16N7400+1💤	0.74	“ misunderstood”“ understanding”“ misunderstanding”
6. L16N1865-1🕒	0.74	“ understanding”“ fundamental”“ fundamentally”
7. L19N2106+1🕒	0.74	“ understanding”“ understand”“ Understanding”
8. L17N7112-1💤	0.74	“ understanding”“ understood”“ knowledge”
9. L16N7064-1🕒	0.73	“ thinking”“认为”“ thinks”
10. L10N2274-1🕒	0.73	“ understanding”“ awareness”“ clarity”

Shadow specimen, apart from the ten: L16N6818-1💤, 0.09.

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 16, neuron 6818, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.63, top-1 agreement 0.68) 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
L16N6818+1 t0-v0.1	0.631	0.68	0.799	1.38	Top boosted tokens ↑
L16N6818-1💤 shadow specimen t0-v0.1	0.587	0.72	0.704	1.00	“ among”“ varied”“ comparable”“ various”“较多”“ parmi”“ diverse”“among”

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.