Layer 5 · Neuron 7298 · boosted

L5N7298+1

Word quirks:

“ players”“ player”“ Pokemon”“ Player”“ RNG”“ gameplay”“ game”“ tournaments”

🕒 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)

Who I Am:
I am a high-velocity, high-impact offensive specialist designed to dismantle the current meta by exploiting the weaknesses of the top-tier decks (specifically Standard and Expanded). My identity is rooted in the belief that I can secure a massive win rate for my team by targeting specific, overplayed strategies. I am not a "safe" card in the traditional sense; rather, I am a "must-play" card that forces opponents to adapt or be outclassed if they do not play me. My design philosophy prioritizes aggressive, low-cost, and high-damage potential over conservative play

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. L5N9189-1🕒	0.94	“ player”“ PvP”“ players”
2. Game Mode✅ L8N4665-1	0.94	“ gameplay”“ players”“ game”
3. L6N5245+1🕒	0.94	“ game”“ gameplay”“ player”
4. The Lobby✅ L3N294+1,L8N4665-1	0.94	“ gameplay”“ players”“ player”
5. The Localizer✅ L3N294-1,L8N4665-1	0.93	“ players”“<\|im_end\|>”“ gameplay”
6. L3N5545-1🕒	0.93	“ gameplay”“ player”“ game”
7. L6N4238+1🕒	0.92	“ players”“ PvP”“ player”
8. L7N9006+1🕒	0.92	“ game”“ gameplay”“ player”
9. L4N8186+1🕒	0.92	“ gamers”“ players”“ game”
10. L7N4685-1🕒	0.92	“ player”“ gameplay”“ Player”

Shadow specimen, apart from the ten: L5N7298-1🕒, -0.07.

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 7298, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.95, top-1 agreement 0.46) 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
L5N7298+1 t0-v0.1	1.949	0.46	2.241	2.24	Top boosted tokens ↑
L5N7298-1🕒 shadow specimen t0-v0.1	1.478	0.54	1.746	1.75	“ solution”“ ‘”“ innovation”“ reference”“ “”“ referenced”“ technology”“ solutions”

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.