Neuron Explorer

Qwen3-0.6B

'to' or 'on' after 'was' or 'were' (planned future action)

A neuron in layer 9 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 small word after 'was', 'were' or 'planned', such as 'was to' or 'plan on', where a future purpose or intention is about to be stated. Turning it up made the model write about plans and purposes, though it often went off track.

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

    against the Raja, who had begun hostilities against the British. Due to the poor state of roads, the force was divided into four columns, which were to enter Coorg from different directions and converge on the capital of Mercara. On

  2. #2 peak

    out its most famous mission in May 1960, undertaken by CIA pilot Francis Gary Powers from an airfield in Pakistan bound for Norway.⏎He was to make a 2,900 mile diversion through Soviet airspace. When he

  3. #3 peak

    boosted on their flight produced its first test beam. The Airborne Laser (ABL) was designed to patrol near the borders of hostile countries. The ABL was to be installed in a modified Boeing 747 for further testing in 2

  4. #4 peak

    Tank Battalion 203 of the 1st Panzer Division, as the PzKpfw 731 (p) and they were to see action in both Norway and France in 1940.⏎Twin

  5. #5 peak

    Strike Fighter (JSF). The contract had a potential total value of over $200 billion during the life of the program. The U.S. planned to buy nearly 3,000 JSFs for its air force, navy

  6. #6 peak

    "twin cities" draw thousands of Mexican people into the already overpopulated region; thus, making the area a major "cesspool". Twin plants were to act as economy boosters for both the US and Mexico; nevertheless, the border

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 sense of which I can see. I can see that it is a number." and then say it out loud, "112, 123, 134, 145,

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

    are able to explore moral consciousness and responsibility, as well as themes of charity and giving. Themes of mental health are also explored through extracts such as ‘The Tell Tale Heart’

  2. #2 peak

    small in the night sky. We sometimes sing the popular nursery rhyme, “Twinkle, twinkle little star….” Actually, however, some of them (like the Sun)

  3. #3 peak

    osporins. Varying degrees of inhibition of the beta-lactamase with clavulanic acid have been reported. Infections due to multidrug-resistant Gram-negative pathogens in the ICU have invoked the re-emergence of col

  4. #4 peak

    o. For this reason, technical training and further education is subject to enormous change.⏎Learning trends in basic and advanced training⏎Many studies show that people's attention span has fallen sharply in recent years. This is

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. L8:60 weakly excites
  2. L8:1073 weakly excites
  3. L7:1148 weakly suppresses
  4. L7:1853 weakly suppresses
  5. L6:323 weakly excites
  6. L8:1330 weakly suppresses
  7. L8:2677 weakly suppresses
  8. L7:71 weakly suppresses
  9. L3:1373 weakly excites 'be' after 'to' in infinitives
  10. L7:851 weakly suppresses
  11. L7:2530 weakly excites
  12. L8:816 weakly excites vague "certain" and "such" before a noun
  13. L4:2817 weakly excites
  14. L6:1830 weakly suppresses
  15. L8:815 weakly excites
  16. L3:606 weakly excites
  17. L3:2887 weakly suppresses
  18. L3:1914 weakly suppresses
  19. L7:2363 weakly suppresses
  20. L8:1312 weakly suppresses
  21. L7:2416 weakly suppresses
  22. L0:2774 weakly suppresses
  23. L5:2107 weakly suppresses
  24. L3:2554 weakly suppresses

Feeds later layers

  1. L14:2188 excites
  2. L18:1017 suppresses
  3. L10:1844 weakly excites
  4. L11:599 weakly excites
  5. L11:2062 weakly excites
  6. L10:707 weakly excites
  7. L15:471 weakly excites
  8. L15:1870 weakly excites
  9. L16:1508 weakly suppresses
  10. L11:2698 weakly suppresses
  11. L10:2694 weakly suppresses
  12. L13:523 weakly suppresses
  13. L13:1400 weakly suppresses "years" after a distance or time number
  14. L23:995 weakly suppresses
  15. L16:2714 weakly excites
  16. L11:38 weakly suppresses
  17. L24:1072 weakly excites
  18. L23:2946 weakly suppresses
  19. L15:1872 weakly suppresses
  20. L23:2652 weakly excites
  21. L19:2327 weakly excites
  22. L17:451 weakly excites
  23. L16:251 weakly suppresses
  24. L18:3005 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.

1 values between -2.44 and -2.323 values between -1.67 and -1.544 values between -1.54 and -1.412 values between -1.41 and -1.293 values between -1.29 and -1.167 values between -1.16 and -1.036 values between -1.03 and -0.9014 values between -0.90 and -0.7718 values between -0.77 and -0.6436 values between -0.64 and -0.5160 values between -0.51 and -0.39128 values between -0.39 and -0.26308 values between -0.26 and -0.131367 values between -0.13 and 0.001388 values between 0.00 and 0.13341 values between 0.13 and 0.2695 values between 0.26 and 0.3933 values between 0.39 and 0.5229 values between 0.52 and 0.6515 values between 0.65 and 0.7710 values between 0.77 and 0.9011 values between 0.90 and 1.038 values between 1.03 and 1.162 values between 1.16 and 1.291 values between 1.42 and 1.553 values between 1.55 and 1.676 values between 1.67 and 1.803 values between 1.80 and 1.931 values between 1.93 and 2.061 values between 2.06 and 2.191 values between 2.19 and 2.321 values between 2.32 and 2.452 values between 2.57 and 2.70offnegativepositiveoften

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. LING
  2. (mod
  3. ULO
  4. simulate
  5. Forever
  6. ins
  7. ulet
  8. 模拟
  9. .SelectCommand
  10. blers

Suppresses

  1. ��
  2. внимание
  3. sehen
  4. 是否存在
  5. hài
  6. :number
  7. xiety
  8. nier
  9. 馁
  10. awy