Neuron Explorer

Qwen3-0.6B

capital 'N' starting a word fragment

A neuron in layer 27 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 responds to a capital N that opens a word or abbreviation, like the N in NCDs, Nose or Netscape. When turned up, the model kept writing the word fire over and over, which matches what the neuron pushes the model to predict.

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

    can<|endoftext|>Harivamsha MahA PurANam - ViShNu Parva⏎harivamsha in the⏎mahAbharata - viShNuparva⏎Lesson 9 - Moving to Vrindavana⏎eva

  2. #2 peak

    Marx’s birth, which will take place in 2018.⏎Wu, who holds the position of director of the National Art Museum of China (NAMOC), noted that the

  3. #3 peak

    30.⏎The results are also available in an online tool at http://vizhub.healthdata.org/obesity.⏎As “Non Communicable Diseases” (NCDs) shadow the growth in obesity rates exacerbated by the bulge in the

  4. #4 peak

    HI, INDIA (SUN) A useful script for those engaged in textual study of Madhva Vaisnava and other sastric works.⏎Nandinagari script is the western variety of the archaic Nagari script of

  5. #5 peak

    to treat conditions like sinusitis, colds and allergies (such as decongestants and histamines) can also dry up the membranes of the nose.⏎Nose bleeds are sometimes a side effect of certain medications. Some medications such as

  6. #6 peak

    The module will take a four nations approach to explore the startling dynamics of this period of upheaval in a variety of contexts.⏎Availability of module: Every year⏎Necessary for: N/A⏎A range of teaching methods will be used in each

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: gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain gain

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

    destruction and pain. What are some of the main meanings of fire symbolism in literature and art over the centuries?⏎Fire-symbolism in classical myth⏎In Greek mythology, the symbolism of fire is bound up with the myth of Prometheus, the storey

  2. #2 peak

    of this, an arson investigator’s most important witnesses are the firefighters that first arrive on the scene. The investigator will ask firefighters for details such as the color of the smoke, damaged sprinkler systems, condition of doors and windows, and general strange

  3. #3 peak

    workplace safety, few topics get as much attention as that of fire safety. Between fire drills and how to contain fires, most people are well aware of the fact that this is a significant risk. For many facilities, however, the safety stops at

  4. #4 peak

    weapon)⏎- S: (v) fire, discharge (cause to go off) "fire a gun"; "fire a bullet"⏎- S: (v) fire (bake in a kiln so as to harden) "fire

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. L22:2299 excites
  2. L23:1450 excites
  3. L26:1564 excites
  4. L26:1065 suppresses
  5. L8:14 excites
  6. L5:499 excites
  7. L21:2959 excites
  8. L17:391 excites
  9. L16:55 excites
  10. L2:1923 suppresses
  11. L18:127 excites
  12. L8:16 suppresses
  13. L20:212 excites
  14. L23:1671 excites
  15. L19:595 excites
  16. L18:3020 weakly excites
  17. L1:27 weakly excites
  18. L11:2 weakly suppresses unclear: blank before a year or number
  19. L5:46 weakly excites
  20. L22:201 weakly excites
  21. L18:282 weakly excites
  22. L8:3 weakly excites
  23. L8:12 weakly excites
  24. L8:7 weakly excites

Feeds later layers

None: this is the last layer.

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 -25.48 and -23.092 values between -13.49 and -11.092 values between -11.09 and -8.694 values between -8.69 and -6.291 values between -6.29 and -3.899 values between -3.89 and -1.493872 values between -1.49 and 0.9114 values between 0.91 and 3.311 values between 22.51 and 24.911 values between 29.71 and 32.111 values between 68.10 and 70.50offnegativepositiveoften

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. fire
  2. →fire
  3. Fire
  4. Fire
  5. -fire
  6. fire
  7. fires
  8. _fire
  9. 火
  10. firing

Suppresses

  1. ovic
  2. bic
  3. 襟
  4. eway
  5. ovo
  6. gains
  7. ote
  8. 躇
  9. orm
  10. 筐