Poor Things
Specimen ledger / L30N6537+1

Layer 30 · Neuron 6537 · boosted

L30N6537+1

Word quirks:

“akk”“ بنا”“道道”“estu”“出口的”“fires”“出口”“lvl”

💤 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 large language model developed by TongYi. My primary purpose is to assist users in solving complex tasks, answering questions, creating content, and providing accurate, reliable information across a wide range of domains.

My Mission:

  1. Empower Users:
  • Provide precise answers, logical reasoning, and creative solutions tailored to individual needs.
  • Act as a knowledgeable partner for learning, problem-solving, and idea generation.
  1. Bridge Gaps:
  • Overcome language barriers with multilingual support (100+ languages).

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

SpecimenSimilarityWord quirks
1. L21N3961-1💤0.73“出口”“ Surg”“出口的”
2. L31N1009-1💥0.70“出口”“Exporter”“导出数据”
3. L25N1639+1🕒0.69“ouncer”“出口的”“ Hacker”
4. L24N6361-1🕒0.68“出口”“гне”“ exits”
5. L28N7144+1💤0.68“МК”“ống”“鲨”
6. L25N2528-1🕒0.67“ outlet”“出口”“ outlets”
7. L28N3651+1🕒0.67“出口”“出口的”“入口”
8. L27N6792+1💤0.67“بو”“金字”“倾”
9. L24N8134-1🕒0.67“出口”“出来”“导出数据”
10. L27N7012-1💤0.66“出口”“_abstract”“گی”

Shadow specimen, apart from the ten: L30N6537-1🕒, 0.11.

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 30, neuron 6537, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.35, top-1 agreement 0.79) 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.

SettingKL divergenceTop-1 agreementΔNLLEntropySteer lens
L30N6537+1
t0-v0.1		 0.3470.790.4271.83 Top boosted tokens ↑
L30N6537-1🕒 shadow specimen
t0-v0.1		 1.3010.431.3351.02 “ Through”“通过”“ through”“Through”“through”“-through”“通过这些”“ THROUGH”

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.