Layer 11 · Neuron 9159 · boosted

L11N9159+1

Word quirks:

“ now”“ previous”“ “”“ ‘”“ later”“ referenced”“ typical”“ previously”

💤 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 Qwen, a large language model developed by Tongyi Lab. My mission is to assist users in completing various tasks, whether they involve answering questions, generating text, or performing other functions. I can help with a wide range of topics, from general knowledge to specialized content, adapting to different contexts and needs. How can I assist you further?

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. Phantom Memoir✅ L5N8933+1	0.86	“ earlier”“ ‘”“ later”
2. L6N5014+1💤	0.80	“ previous”“ previously”“ recently”
3. L9N7037+1💤	0.79	“ before”“ ‘”“ now”
4. L11N3422-1🕒	0.79	“ sometimes”“ **“”“ now”
5. L8N4337-1💤	0.79	“ now”“ previously”“ recently”
6. L15N794+1🕒	0.79	“ earlier”“ previous”“ your”
7. L11N2473+1🕒	0.78	“ info”“ previous”“ ‘”
8. L10N640+1🕒	0.78	“ referenced”“ now”“ previous”
9. L14N1026+1🕒	0.78	“ previous”“ earlier”“ previously”
10. L11N2463+1💤	0.77	“ previously”“ previous”“ earlier”

Shadow specimen, apart from the ten: L11N9159-1🕒, 0.15.

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 9159, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.79, 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
L11N9159+1 t0-v0.1	0.787	0.68	1.029	1.20	Top boosted tokens ↑
L11N9159-1🕒 shadow specimen t0-v0.1	1.064	0.62	1.383	1.62	“ neither”“ control”“ unknown”“ no”“ yet”“ controlled”“ controls”“ nonetheless”

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.