Layer 6 · Neuron 4877 · boosted

L6N4877+1

Word quirks:

“ **“”“ **”“ **‘”“ **"”“ **«”“ **'”“ **„”“ "**”

💤 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 Qwen3.5, the latest iteration in the Tongyi series of large language models developed by Alibaba. My mission is to empower users with advanced capabilities across a wide range of tasks, from generating high-quality content to solving complex problems.

Key Capabilities:

Enhanced Reasoning: I excel at handling intricate logical reasoning tasks, such as mathematical proofs, scientific simulations, and strategic planning.
Multilingual Support: Fluent in over 100 languages, enabling seamless communication for global audiences.
Creative Content Generation: Specialized in crafting engaging narratives, articles,

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. L9N2895+1💤	1.00	“ "”“ “”“ **”
2. L15N6038-1💤	1.00	“ “”“ "”“ **”
3. L16N5971-1💤	0.99	“ "”“ '”“ **“”
4. L16N665+1💤	0.99	“ “”“ "”“ **”
5. L15N9086-1🕒	0.99	“ “”“ "”“ **”
6. L10N7836-1💤	0.99	“ ”“ “”“ **"”
7. L17N309+1🕒	0.99	“ "”“ “”“ **”
8. L16N8419+1🕒	0.99	“ "”“ '”“ **”
9. L10N6183-1🕒	0.99	“ "”“ "”“ **“”
10. L6N6499-1💤	0.99	“ "”“ '”“ **“”

Shadow specimen, apart from the ten: L6N4877-1🕒, 0.27.

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 6, neuron 4877, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.56, top-1 agreement 0.72) 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
L6N4877+1 t0-v0.1	0.556	0.72	0.669	1.31	Top boosted tokens ↑
L6N4877-1🕒 shadow specimen t0-v0.1	2.489	0.38	2.917	2.80	“ aforementioned”“ ____”“ _____”“<\|im_end\|>”“ referenced”“ said”“ entity”“ refer”

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.