Neuron Explorer

Qwen3-0.6B

'question' after 'this'

A neuron in layer 0 of Qwen3-0.6B. Its strongest snippets are very consistent: they mostly peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on the word 'question' when it follows 'this', as in 'this question'. It is a narrow pattern. Turning it up produced repetitive, unrelated text, so the steering shows nothing about the pattern.

Written by Claude from this neuron's activations and steering outputs. Confidence: medium. Think it is wrong? Flag it (opens a prefilled issue on GitHub).

Fires most on

The text windows where this neuron's activation was highest. Amber marks positive activation; the outlined token is the peak. Hover a token for its strength.

  1. #1 peak

    this question by reviewing the scientific literature on cow-calf separation to provide perspective and, hopefully

  2. #2 peak

    generalization.)⏎Your reader will be thankful if you start with “This question…” or “This wig…” or, most likely, “This intrusion…” If

  3. #3 peak

    and less justifiable answers. Something like “Which of the following motivations best explains so-and-so’s actions in such-and-such novel?” Instead of posing this question, hoping that all your students take a moment to think about it, then hearing

  4. #4 peak

    2021, Fox News report, Tucker Carlson asked the question no one is really allowed to ask: “How many Americans have died after taking

  5. #5 peak

    ? This question has now been answered thanks to Read more…⏎Tag: ancient inscription⏎12

  6. #6 peak

    motion and enterprise.⏎This Elibron Classics publication is a facsimile reprint of a 1901 variation by means of the Macmill<|endoftext|>This question seems like such a common occurrence. It most always follows the realization that we have

Steering

Push this neuron up or down while the model writes, and watch what changes. The outputs were generated ahead of time, so nothing runs in your browser.

turned down a lot
Prompt

Here is something I have been thinking about lately: I am going, and I and the? I am...? I am going to...? I am going to...? I am going to...? I am going to...? I am going to...? I am going to...

Most negative on

Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.

  1. #1 peak

    the Cuban government of Batista. Subsequently, he became an important figure in Cuban diplomacy and a vocal critic of the United States and the Soviet Union. Later on he helped revolutionary groups in Congo and Bolivia until he was captured and executed by the

  2. #2 peak

    except one guy. The guy is named Bao Shuya, who was the one saved his ass and took him to Ju as a political asylum seeker and later on helped him regain his lordship. Duke’s displeased and asked why he’s

  3. #3 peak

    of feudal lords in China in the Eastern state of Qi, took shelter in the state of Ju when he was persecuted in a bloody coup by his uncle. Later on another coup took place over Qi’s court and he got to

  4. #4 peak

    story celebrates the important and impressive feat of a small but mighty creature.⏎Christy Ottaviano Books.⏎Hardcover Tiny Bird download PocketBook on Kobo.

Connections

Show on map

Which neurons in earlier layers feed this one, and which later ones it feeds. Amber links excite, blue links suppress; longer bars are stronger.

Fed by earlier layers

None: this is the first layer.

Feeds later layers

  1. L5:2036 weakly excites
  2. L6:2668 weakly suppresses
  3. L4:446 weakly suppresses
  4. L5:202 weakly excites
  5. L17:1784 weakly suppresses
  6. L19:2031 weakly excites line breaks in web footers and metadata
  7. L14:1256 weakly excites
  8. L2:2040 weakly suppresses
  9. L9:1727 weakly suppresses
  10. L3:2440 weakly suppresses
  11. L13:42 weakly suppresses
  12. L7:2582 weakly suppresses
  13. L12:1260 weakly suppresses
  14. L21:161 weakly excites
  15. L2:17 weakly suppresses
  16. L19:412 weakly excites
  17. L5:1715 weakly excites
  18. L12:1542 weakly suppresses
  19. L1:1710 weakly suppresses
  20. L15:839 weakly suppresses
  21. L9:1352 weakly suppresses
  22. L17:1168 weakly excites
  23. L14:67 weakly excites
  24. L14:2834 weakly excites

Activation histogram

How often this neuron fires at each strength. Almost always it does close to nothing (the tall bars in the middle); the snippets above are the rare moments it fired hard, amber for positive and blue for negative.

2 values between -0.41 and -0.371 values between -0.37 and -0.332 values between -0.33 and -0.297 values between -0.29 and -0.2420 values between -0.24 and -0.2034 values between -0.20 and -0.1677 values between -0.16 and -0.12174 values between -0.12 and -0.08399 values between -0.08 and -0.041533 values between -0.04 and 0.001023 values between 0.00 and 0.04315 values between 0.04 and 0.09158 values between 0.09 and 0.1351 values between 0.13 and 0.1730 values between 0.17 and 0.2115 values between 0.21 and 0.2515 values between 0.25 and 0.2915 values between 0.29 and 0.334 values between 0.33 and 0.374 values between 0.37 and 0.4113 values between 0.41 and 0.463 values between 0.46 and 0.504 values between 0.50 and 0.542 values between 0.54 and 0.581 values between 0.58 and 0.622 values between 0.62 and 0.661 values between 0.83 and 0.871 values between 0.91 and 0.951 values between 0.95 and 0.991 values between 1.20 and 1.24offnegativepositiveoften

Logit lens

The words this neuron nudges the model toward (amber) or away from (blue) when it fires. A hint, not a label: many neurons work through later layers instead.

Promotes

  1. depend
  2. depends
  3. ependency
  4. 蝶
  5. 幻
  6. extrad
  7. -fe
  8. pest
  9. cosmetic
  10. tail

Suppresses

  1. 留言板
  2. том
  3. 录制
  4. acular
  5. ocalypse
  6. INCLUDED
  7. orial
  8. idges
  9. missão
  10. 服务能力