Neuron Explorer

Qwen3-0.6B

end of document marker

A neuron in layer 15 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 mostly reacts to the invisible marker that marks the end of one text and the start of an unrelated one. When it was turned up, the model's writing got stranger and less on topic, so its role seems to be about document boundaries rather than a clear topic.

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

    cleaner fuel (good for immediately reducing greenhouse gases) and one that buys offsets for planting trees (not as good) that might be cut down a few years later<|endoftext|>Hand-coloured lithograph entitled; 'Dampfmaschinen Pfer

  2. #2 peak

    fur is dark brown and they have white tuft of hair on the top of their heads. Cotton Top Tamarins live for about 16 years.<|endoftext|>When a pupil has a hard time writing their papers, it is typical for them

  3. #3 peak

    being caused by curve space when neither were in motion, what causes the force to act upon each other.⏎Hope this is clear enough to get my question across.<|endoftext|>Yesterday I cut myself some fresh pineapple, then promptly proceeded to

  4. #4 peak

    more. According to another computer model, organic farmers are expected to net an average of $50 to $60 more per acre than their conventional counterparts.<|endoftext|>It is sometimes felt that saving energy by natural ventilation in atria conflicts

  5. #5 peak

    between AD1260 and 1390.⏎However, many scholars have provided multiple scientific reasons why the radiocarbon dating is not reliable.<|endoftext|>The Smithsonian Magazine offers a brief history lesson regarding gun control:⏎Marshall Virgil

  6. #6 peak

    Marx has been imprinted in his heart since he was a child, adding that “Marx molded by a Chinese artist contains emotion and [warmth].”<|endoftext|>Citadel of Herat⏎This citadel, which has been suggested as another

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: The the the, which is the. I. The, which is the. For the: (1. This: [the, in. (2. The. (4.3.3.3.3.3.3

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

    smooth variations which are precise to the unique paintings.⏎Sextus Empiricus' Outlines of Pyrrhonism is among the most crucial and influential texts within the historical past of Greek philosophy. within the calls for of cause Casey Perin

  2. #2 peak

    desires, the right kind method to pursue happiness, and the correct plan for an excellent lifestyles.⏎Read or Download Aristotle for Everybody: Difficult Thought Made Easy PDF⏎Best greek &

  3. #3 peak

    the lesion. Collectively, DNA restoration pathways keep up with the fidelity from the human being genome. Open up in another window Shape 1 Brief Overview of DNA Damage Repair: Three common types

  4. #4 peak

    subtractions are performed, by way of example, or how to identify the visible difference between a optimistic number and a unfavorable number. By incorporating activities that strengthen recent abilities with brand new ones, you can strengthen your child’s basic abilities to make

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

  1. L10:1255 excites
  2. L10:689 excites
  3. L10:428 excites
  4. L9:483 excites
  5. L3:149 excites
  6. L10:1165 excites
  7. L11:308 excites
  8. L10:507 suppresses
  9. L10:348 suppresses
  10. L3:791 excites
  11. L10:2625 excites
  12. L7:1333 excites
  13. L11:2388 weakly suppresses
  14. L9:511 weakly suppresses
  15. L9:2581 weakly suppresses
  16. L14:180 weakly excites
  17. L10:493 weakly suppresses
  18. L9:2618 weakly excites
  19. L10:196 weakly suppresses
  20. L14:523 weakly excites
  21. L12:1132 weakly excites
  22. L11:1491 weakly excites
  23. L3:117 weakly suppresses
  24. L6:13 weakly excites

Feeds later layers

  1. L16:34 excites
  2. L16:97 excites
  3. L16:1040 excites
  4. L16:83 excites
  5. L16:28 excites
  6. L16:38 excites
  7. L16:31 excites
  8. L16:126 excites
  9. L16:8 excites
  10. L16:73 excites
  11. L16:9 excites
  12. L17:916 suppresses
  13. L16:39 excites
  14. L16:90 excites
  15. L16:188 excites
  16. L16:100 excites
  17. L16:107 excites
  18. L16:0 excites
  19. L16:119 excites
  20. L16:2461 excites
  21. L16:2 excites
  22. L16:25 excites
  23. L16:137 excites
  24. L16:19 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.

1 values between -2.15 and -2.041 values between -2.04 and -1.941 values between -1.43 and -1.332 values between -1.23 and -1.133 values between -0.92 and -0.821 values between -0.82 and -0.723 values between -0.72 and -0.628 values between -0.62 and -0.5116 values between -0.51 and -0.4143 values between -0.41 and -0.31117 values between -0.31 and -0.21226 values between -0.21 and -0.11465 values between -0.11 and -0.00725 values between -0.00 and 0.10687 values between 0.10 and 0.20641 values between 0.20 and 0.30415 values between 0.30 and 0.40261 values between 0.40 and 0.51151 values between 0.51 and 0.6180 values between 0.61 and 0.7133 values between 0.71 and 0.8120 values between 0.81 and 0.914 values between 0.91 and 1.021 values between 1.22 and 1.321 values between 1.73 and 1.832 values between 1.83 and 1.93offnegativepositiveoften

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. lict
  2. ượ
  3. 万象
  4. 以人为
  5. 交代
  6. stag
  7. 以色
  8. 最低
  9. wegen
  10. 口径

Suppresses

  1. ionate
  2. ions
  3. IENCE
  4. cion
  5. cation
  6. ’all
  7. ampions
  8. ials
  9. favourite
  10. ERNEL