Neuron Explorer

Qwen3-0.6B

"to" before an infinitive verb

A neuron in layer 6 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 lights up on the word "to" when it comes right after a word like "important," "successful" or "took it upon himself" and is about to introduce an action, as in "important to make sure." When it was turned up, the model's replies did not show a 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

    irrigation through early plant development (several weeks), will be necessary in our dry summer climate. In shad areas, like the one above, its more successful to plant out roots stalk instead of seeding, but I always throw something down, just

  2. #2 peak

    .”⏎Since the IT band’s primary function is to stabilize the pelvis, thigh, and knee during movement, its important to make sure your muscles are strong enough to do so. Dr. Metzl added

  3. #3 peak

    that the symphony is dead… but when the looked-for genius comes, it may be absolutely revived.”⏎A few years later, Elgar took it upon himself to be the “looked-for genius,” and the result was a staggering immediate success

  4. #4 peak

    We know that it is consistent to be able to live and not know. Some people say, "How can you

  5. #5 peak

    It is fascinating to find out how you came to be on this earth and who lived before you in

  6. #6 peak

    dimension to help others, give of one’s time, resources and energy, in aid of others.⏎- The Regulatory Dimension: The Regulatory drive indicates one’s drive to establish order, routine and structure. This motivation is to promote rules and policies,

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 have a very specific example of the last one I have, I have a very specific example of the last one I have... I have a very specific example of the last one I have... I have a very specific example of the last

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

    have no underlying health issues, such as diabetes,” says Cohen. “And if it promotes weight loss, that can have great psychological benefits, which may spur you on to other healthy behaviors such as making regular exercise part of your life, too.”⏎

  2. #2 peak

    drinking alcohol.⏎Alcohol withdrawal occurs most often in adults. But, it may also occur in teenagers or children.⏎The more you drink regularly, the more likely you are to develop alcohol withdrawal symptoms when you

  3. #3 peak

    top causes of discoloration include food and drinks that have deep color pigmentation. in any food, the deeper or more intense the color, the more likely it will stain teeth. Teas, wines, soft drinks, sports drinks, berries

  4. #4 peak

    because fire ants prefer monocultures and will shy away from biodiverse areas.⏎By contrast, the more kinds of species that inhabit an area, the more likely it is that at least a few strains of plants or animals will be resistant to pests

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. L5:2742 excites
  2. L4:2908 excites
  3. L5:1885 excites
  4. L5:2744 excites
  5. L4:635 weakly suppresses
  6. L4:2577 weakly suppresses
  7. L4:758 weakly excites
  8. L2:2769 weakly suppresses
  9. L4:398 weakly excites
  10. L5:102 weakly excites
  11. L1:949 weakly excites 'to' after 'agreed'
  12. L4:2308 weakly excites
  13. L5:110 weakly suppresses
  14. L4:1819 weakly excites
  15. L4:2470 weakly suppresses
  16. L2:2798 weakly suppresses
  17. L5:86 weakly suppresses
  18. L4:944 weakly suppresses
  19. L3:2614 weakly suppresses
  20. L4:1316 weakly suppresses
  21. L5:2217 weakly suppresses
  22. L5:1950 weakly excites unclear: 'Lorenz' and name fragments
  23. L2:1163 weakly suppresses
  24. L3:1457 weakly excites

Feeds later layers

  1. L8:369 strongly suppresses 'of' after 'learned' or 'because'
  2. L8:94 strongly excites
  3. L16:1153 strongly excites
  4. L13:909 strongly excites 'to' before a verb after a noun
  5. L7:681 excites 'from' after prevent or stop
  6. L15:1176 excites
  7. L9:226 excites 'to' starting an infinitive
  8. L19:385 suppresses
  9. L18:472 excites
  10. L7:138 excites 'to' after verbs like prove or time
  11. L8:723 excites
  12. L14:659 excites
  13. L18:715 excites
  14. L17:1592 excites
  15. L19:617 excites
  16. L17:472 excites
  17. L19:776 excites
  18. L20:480 excites
  19. L17:187 excites
  20. L17:1994 excites
  21. L7:8 suppresses
  22. L19:652 excites
  23. L19:2515 excites
  24. L20:241 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 -3.25 and -3.031 values between -1.50 and -1.282 values between -1.28 and -1.064 values between -1.06 and -0.846 values between -0.84 and -0.6224 values between -0.62 and -0.40243 values between -0.40 and -0.182277 values between -0.18 and 0.041231 values between 0.04 and 0.2676 values between 0.26 and 0.4714 values between 0.47 and 0.691 values between 0.69 and 0.914 values between 0.91 and 1.134 values between 1.13 and 1.355 values between 1.35 and 1.572 values between 1.57 and 1.792 values between 1.79 and 2.015 values between 2.01 and 2.231 values between 2.88 and 3.103 values between 3.10 and 3.321 values between 3.54 and 3.761 values between 5.29 and 5.51offnegativepositiveoften

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. 把它
  2. /Images
  3. ��
  4. 将其
  5. ivid
  6. 把她
  7. 론
  8. stalk
  9. manufacture
  10. ifice

Suppresses

  1. weakest
  2. 不确定性
  3. gameState
  4. edido
  5. eso
  6. principio
  7. consistency
  8. etas
  9. ewitness
  10. ::~