Neuron Explorer

Qwen3-0.6B

'to' after a verb or noun

A neuron in layer 5 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

The neuron lights up on the small word 'to' when it follows a verb or noun, as in 'put to use', 'right to recall' or 'went to live'. Turning it up made the model's writing drift into loose, repetitive or off-topic text, with no clear theme.

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

    the 19th cent. when Pasteur noted the antagonistic effect of other bacteria on the anthrax organism and pointed out that this action might be put to therapeutic use. Meanwhile the German chemist Paul Ehrlich developed the idea

  2. #2 peak

    59; died at Madrid, 27 March, 1625. He was a great-grandson of the Tordesillas who was put to death by the Comuneros at Seville. He studied in Spain and Italy

  3. #3 peak

    istic reforms and translated the Bible into his native tongue.⏎Many Poles turned from the Roman Church, winning a measure of religious freedom. Some church buildings were put to Protestant use. Laws forbade the execution of Roman Church decrees

  4. #4 peak

    implemented the law of giving a right to recall to the employers of certain industries. This law has been enacted until the end

  5. #5 peak

    of the narrative “My first mistress was a woman of… To “l had resided but a short time in Baltimore” to “Very soon after I went to live with,” Douglass shows to his readers a clear time pattern, his whereabouts

  6. #6 peak

    like Pope Francis who at the World Day for Youth in Rio de Janeiro, Brazil, called for the "respect and protection of the Creation which God has confided to humankind," As educators, the Sisters read, research, study and teach about

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 will say, for example, that the value of the expression is 2000. I will say, for example, that the value of the expression is 5. I will say, for example, that the value of

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

    maintained a normal weight. (The reason for this difference remains a mystery.)⏎Cowley's team then removed the hypothalamus from the animals and analysed the response of cells from this brain region to leptin. When exposed to the hormone, the

  2. #2 peak

    Japanese public opinion and Japan–South Korea relations.⏎An invaluable resource for scholars of East Asian regional security issues.⏎Table of Contents⏎- Bringing People’s Voices Back in Japan–South Korea Relations⏎- South Korea-Japan FCR Crisis and Public

  3. #3 peak

    This was mainly because of the Friar.⏎The Capulets and Montagues were not on good terms with each other. Unfortunately though, Romeo and Juliet fell in love with each other regardless if they are from opposing families. Because of these families

  4. #4 peak

    men heard the shells edging closer. “And then bang!—one of the big shells struck the ground right in front of us and we all went up in the air. But we all come down again.”⏎York was showered with honors

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. L0:2060 excites
  2. L4:2373 excites
  3. L0:1956 weakly suppresses
  4. L4:561 weakly excites
  5. L2:3058 weakly suppresses
  6. L4:1568 weakly suppresses
  7. L1:2801 weakly excites
  8. L4:1532 weakly suppresses
  9. L4:2649 weakly excites
  10. L4:1658 weakly excites
  11. L0:1027 weakly excites
  12. L4:1880 weakly suppresses
  13. L4:2122 weakly excites
  14. L2:1060 weakly excites
  15. L3:589 weakly suppresses
  16. L0:1217 weakly excites
  17. L3:799 weakly suppresses
  18. L4:1990 weakly excites
  19. L4:2146 weakly excites
  20. L1:750 weakly suppresses "of" after "kind" or "sort"
  21. L2:977 weakly excites
  22. L2:2423 weakly suppresses
  23. L3:2878 weakly suppresses
  24. L2:1157 weakly suppresses

Feeds later layers

  1. L6:338 suppresses
  2. L8:1627 excites
  3. L9:723 excites
  4. L8:45 excites
  5. L16:1168 excites 'to' after verbs of change
  6. L12:1851 excites hyphen inside dates
  7. L17:2821 excites
  8. L19:1345 excites
  9. L17:2527 weakly excites
  10. L19:858 weakly excites
  11. L10:132 weakly excites
  12. L10:2213 weakly excites 'to' or 'into' showing movement between two things
  13. L18:1001 weakly excites
  14. L7:14 weakly excites
  15. L10:297 weakly excites
  16. L13:2521 weakly suppresses
  17. L18:1857 weakly excites
  18. L6:1004 weakly excites
  19. L26:524 weakly suppresses
  20. L15:675 weakly suppresses
  21. L16:623 weakly excites
  22. L19:1963 weakly excites
  23. L6:368 weakly suppresses
  24. L10:272 weakly suppresses

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 -1.34 and -1.214 values between -1.07 and -0.944 values between -0.81 and -0.6711 values between -0.67 and -0.5418 values between -0.54 and -0.4156 values between -0.41 and -0.27237 values between -0.27 and -0.141233 values between -0.14 and -0.011849 values between -0.01 and 0.13383 values between 0.13 and 0.2668 values between 0.26 and 0.398 values between 0.39 and 0.5311 values between 0.53 and 0.666 values between 0.66 and 0.793 values between 0.79 and 0.934 values between 0.93 and 1.062 values between 1.46 and 1.592 values between 1.59 and 1.733 values between 1.73 and 1.862 values between 1.86 and 1.991 values between 1.99 and 2.131 values between 2.13 and 2.261 values between 3.86 and 3.99offnegativepositiveoften

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. ppers
  2. pper
  3. gether
  4. 了他的
  5. iration
  6. .githubusercontent
  7. Brow
  8. olini
  9. gal
  10. eing

Suppresses

  1. пед
  2. -validate
  3. upe
  4. хват
  5. nels
  6. фик
  7. 要做
  8. 万美元
  9. pipe
  10. 扉