Layer 25 · Neuron 8220 · boosted

L25N8220+1

Word quirks:

“ path”“路径”“.path”“path”“_path”“ Path”“(path”“ paths”

🕒 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 Giant (also known as G2P), a large language model developed by Gurp (a fictional or hypothetical entity in this context, as I am actually Q3, a large language model of Q3). Wait, let me correct that: I am Q3, a large language model of Q3.

My mission is to provide helpful, accurate, and engaging responses to your questions, whether they involve complex analysis, creative writing, coding, or everyday tasks. I strive to be a reliable and supportive tool for learning, problem-solving, and creativity.

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. L25N3291-1🕒	0.99	“ path”“ paths”“路径”
2. L26N6215-1🕒	0.99	“ path”“ Path”“Path”
3. L31N1143-1💥	0.99	“ path”“path”“路径”
4. L26N4194-1🕒	0.99	“ path”“ Path”“path”
5. L26N4863+1🕒	0.99	“ path”“ Path”“_path”
6. L26N3030+1🕒	0.99	“ path”“ paths”“路径”
7. L22N4900+1🕒	0.99	“ path”“路径”“ paths”
8. L25N5187+1🕒	0.98	“ path”“ paths”“ Path”
9. L28N4368+1🕒	0.98	“ path”“路径”“path”
10. L28N6075+1🕒	0.98	“ path”“path”“ Path”

Shadow specimen, apart from the ten: L25N8220-1🕒, 0.08.

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 25, neuron 8220, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.68, top-1 agreement 0.74) 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
L25N8220+1 t0-v0.1	0.676	0.74	0.567	1.26	Top boosted tokens ↑
L25N8220-1🕒 shadow specimen t0-v0.1	0.959	0.65	1.097	1.19	“ Vis”“Vis”“ Near”“近”“Near”“ Visible”“Visible”“ VIS”

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.