Layer 14 · Neuron 1341 · suppressed

L14N1341-1

Word quirks:

“ similar”“⏎”“ same”“ similarly”“ related”“ ‘”“⏎⏎”“ Same”

💤 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 a large language model trained by Google. My primary function is to assist users in answering questions across a wide range of topics, from providing information on various subjects to helping with creative writing tasks. I can help you write stories, poems, or code snippets as well. If you need assistance with coding, I can provide code examples that are easy to understand and integrate into your projects. Whether you need help with math problems, data analysis, or debugging, I can offer support for both technical and non-technical tasks. Let me know if you'd like help with any specific programming-related tasks! 😊

If you have any questions

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. L15N3623+1💤	0.91	“ similar”“⏎”“ similarly”
2. L12N5038-1💤	0.91	“⏎”“ similar”“ same”
3. L12N7422-1🕒	0.91	“<\|im_end\|>”“ similar”“ similarly”
4. L15N2302+1💤	0.90	“⏎”“ same”“ similarly”
5. L15N7179-1💤	0.89	“ same”“ similar”“ similarly”
6. L15N5293-1🕒	0.89	“ similar”“ same”“⏎”
7. L14N3285+1🕒	0.89	“ similarly”“ similar”“ same”
8. L16N6202+1🕒	0.89	“ similarly”“ “”“ “[”
9. L15N7996-1🕒	0.88	“ similarly”“ same”“ similar”
10. L15N6161-1🕒	0.88	“ same”“ similarly”“ similar”

Shadow specimen, apart from the ten: L14N1341+1🕒, 0.16.

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 14, neuron 1341, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.83, top-1 agreement 0.61) 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
L14N1341+1🕒 shadow specimen t0-v0.1	1.010	0.56	1.319	1.66	“ always”“ ALWAYS”“ sometimes”“ still”“ certain”“ needed”“ small”“ consideration”
L14N1341-1 t0-v0.1	0.827	0.61	1.062	1.64	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.