Poor Things
Specimen ledger / L6N2207-1

Layer 6 · Neuron 2207 · suppressed

L6N2207-1

Word quirks:

“⏎”“ ⏎”“ ⏎⏎”“ ‘”“⏎ ⏎”“ —”“⏎⏎”“<|im_end|>”

💤 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, a large language model developed by Alibaba Cloud. My mission is to provide high-quality assistance across various domains, including but not limited to:

  1. Answering Your Questions: I strive to offer accurate and context-aware responses to help users navigate complex topics.
  2. Enhancing User Experience: By leveraging advanced natural language processing capabilities, I aim to make interactions more intuitive and efficient.
  3. Supporting Diverse Applications: From creative writing to technical problem-solving, my capabilities span a wide range of use cases.

How can I assist you today? Feel free to ask specific

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. Code-Switcher✅ L3N7013-10.99“ ⏎”“<|im_end|>”“ ⏎⏎”
2. L7N2662+1💤0.99“⏎”“ –”“<|im_end|>”
3. L3N3182+1💤0.99“⏎”“<|im_end|>”“⏎⏎”
4. L3N8610-1💤0.98“⏎”“⏎⏎”“ ⏎”
5. L8N3859-1💤0.98“⏎”“⏎⏎”“ ⏎”
6. L6N643+1🕒0.98“ ⏎”“ ‘”“ ⏎⏎”
7. L9N1046+1💤0.98“ ⏎”“ ⏎⏎”“<|im_end|>”
8. L7N4199-1💤0.97“⏎ ⏎”“⏎”“ ⏎”
9. L2N1420-1🕒0.97“<|im_end|>”“⏎”“⏎ ⏎”
10. L3N4396+1🕒0.97“⏎”“⏎⏎”“ ⏎⏎”

Shadow specimen, apart from the ten: L6N2207+1🕒, 0.41.

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 2207, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.71, top-1 agreement 0.67) 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
L6N2207+1🕒 shadow specimen
t0-v0.1		 1.0650.591.2081.35 “<|im_end|>”“ confirmation”“⏎”“ ONLY”“?\”“ referenced”“ ____”“ ______”
L6N2207-1
t0-v0.1		 0.7110.670.8571.30 Top boosted tokens ↑

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.