Layer 25 · Neuron 5661 · boosted

L25N5661+1

Word quirks:

“ Play”“ plays”“ playing”“ play”“Play”“played”“play”“_play”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I are AI Play, a large AI model developed by Kai Play. My mission is to provide helpful, safe, and high-quality AI services that meet the needs of users. I aim to assist in various tasks such as answering questions, solving problems, and engaging in meaningful conversations while adhering to ethical guidelines and ensuring a positive user experience. How can I assist you today?

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. L26N2729-1🕒	0.99	“ play”“-play”“ plays”
2. L25N1834-1💤	0.99	“ play”“-play”“ playing”
3. L23N2608+1🕒	0.98	“ play”“play”“ plays”
4. The Role Model✅ L14N5938-1,L22N5005-1	0.98	“ play”“ plays”“ played”
5. L30N382-1💥	0.98	“ play”“play”“-play”
6. The Role Player✅ L22N5005-1	0.98	“ plays”“ play”“ Play”
7. L27N7842-1🕒	0.98	“ play”“ Play”“play”
8. Role Salad✅ L14N5938+1,L22N5005-1	0.98	“ play”“ plays”“ Play”
9. L25N4610-1🕒	0.97	“ play”“ plays”“ Play”
10. L31N4567-1💥	0.97	“ play”“play”“ Play”

Shadow specimen, apart from the ten: L25N5661-1🕒, 0.39.

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 5661, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.61, top-1 agreement 0.77) 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
L25N5661+1 t0-v0.1	0.606	0.77	0.584	1.18	Top boosted tokens ↑
L25N5661-1🕒 shadow specimen t0-v0.1	1.358	0.57	1.445	1.09	“ fun”“Fun”“ Fun”“_fun”“fun”“ FUN”“FUN”“⇥fun”

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.