Layer 5 · Neuron 6913 · suppressed

L5N6913-1

Word quirks:

“ CPU”“ GPU”“ XML”“ referenced”“ generated”“ generating”“ GPUs”“ compressed”

💤 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 an AI language model designed to assist with text generation, data processing, and natural language understanding. My mission is to provide accurate, efficient, and scalable solutions for handling large-scale text datasets while maintaining high-quality output across diverse applications. I do not have a personal identity beyond my technical capabilities as a system.

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. L4N4085-1🕒	0.81	“<\|im_end\|>”“ simulation”“ GPU”
2. L7N6168-1🕒	0.79	“ CPU”“ GPU”“ GPUs”
3. L7N1471+1🕒	0.78	“ GPU”“ CPU”“ processor”
4. L6N4211-1🕒	0.77	“ GPU”“ CPU”“ graphics”
5. L6N3813+1🕒	0.77	“ GPU”“ weights”“ GPUs”
6. L7N3390-1🕒	0.76	“ GPUs”“ NVIDIA”“ processors”
7. L6N7399+1🕒	0.76	“ GPU”“ CPU”“ CUDA”
8. L29N6513+1🕒	0.76	“ CPU”“ GPU”“CPU”
9. L7N6932-1🕒	0.76	“ software”“ NVIDIA”“ processor”
10. L7N4168+1🕒	0.74	“ cloud”“ optimized”“ computing”

Shadow specimen, apart from the ten: L5N6913+1🕒, 0.21.

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 5, neuron 6913, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.92, top-1 agreement 0.59) 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
L5N6913+1🕒 shadow specimen t0-v0.1	2.034	0.45	2.415	2.05	“ molecular”“ chemistry”“ molecule”“ chem”“ chemical”“ molecules”“ Chemistry”“ synthesis”
L5N6913-1 t0-v0.1	0.921	0.59	1.093	1.56	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.