Neuron Explorer

Qwen3-0.6B

unclear: 'ardu' and 'axi' word fragments

A neuron in layer 2 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron is unclear. It reacts to pieces of longer words, especially 'ARDUINO', 'arduous' and 'axial' or 'axiomatic'. Turning it up or down did not give a clear effect.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

    are checked to verify they meet the required specifications.⏎<|endoftext|>RASPBERRY Pi - LEARNING HOW TO MAKE YOUR ROBOT WORK⏎ANTICS - ARDUINO - ARMOUR - ARTWORK - BIOLOGY - BLACK BOX - COMPUTERS

  2. #2 peak

    slightly more sustainable in the reduction of energy and time used by humans to create robots for a better world. KISS⏎ABOUT RASPBERRY Pi and ARDUINO COMPUTERS⏎The Arduino is a wonderful platform for handling the lower level work

  3. #3 peak

    particular value in ornamental planting. By placing three or five plants in clump-fashion, the beauty of this Broom is greatly enhanced.⏎C. ALDUS INCARNATUS (syn C. incarnatus) resembles C. purpureus

  4. #4 peak

    σχῆμα τῷ λόφῳ in Einklang zu bringen, ist es unumgänglich nothwendig dass der Ausdruck Roma quadrata technisch in doppeltem Sinne gebraucht wurde

  5. #5 peak

    of those very mountains. The tortuous, switchback road with steep climbs, hairpin bends and often cut by bad weather conditions made travel so long and arduous to Palma, that Sóller looked more to Barcelona – a sea

  6. #6 peak

    length and can vary in color. Mating can take from a few hours to a few days with the male sitting atop the female.<|endoftext|>Even when ‘Abdu’l-Bahá strongly disagreed with the religious beliefs from Roman Catholics. Examples

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 need to be able to do this. I need to be able to do this. I need to be able to do this. I need to be able to do this. I need to be able to do this. I need to

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

    - The foundations of analysis, which begins with the study of functions and their derivatives and integrals⏎- The fundamentals of advanced algebra, which is based on axiomatic systems involving operations of addition and multiplication in general settings⏎- Additional subjects such

  2. #2 peak

    dorsum)⏎- Lateral digital vein (near dewclaws)⏎How does infection spread from the coffin joint, and what does it cause?⏎- axially: footrot⏎- abaxially: vertical fissure aka sand crack⏎

  3. #3 peak

    - More weight on the front half⏎What is the blood supply to the claw?⏎- dorsal: dorsal metatarsal a. -> axial proper digital a. -> terminal arch⏎- plantar: terminal arch & marginal

  4. #4 peak

    fasts⏎small, inconspicuous, or lacking. Commonly yellowish, but deep red⏎Medulla appear parenchymatous around central axial cell;⏎cortical filaments with few divisions on radii, outer layer

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. L0:1224 suppresses
  2. L0:178 suppresses
  3. L0:1301 suppresses
  4. L0:1 weakly suppresses
  5. L1:1464 weakly excites
  6. L1:2799 weakly excites
  7. L1:1717 weakly excites
  8. L1:146 weakly excites
  9. L1:6 weakly excites
  10. L0:397 weakly suppresses
  11. L1:262 weakly suppresses
  12. L0:1813 weakly excites
  13. L1:1960 weakly excites
  14. L1:56 weakly excites
  15. L0:495 weakly suppresses
  16. L1:30 weakly excites
  17. L1:1692 weakly excites
  18. L0:644 weakly excites
  19. L1:1583 weakly excites
  20. L1:1757 weakly suppresses
  21. L1:2318 weakly suppresses
  22. L1:3038 weakly excites
  23. L1:2444 weakly suppresses
  24. L1:491 weakly excites

Feeds later layers

  1. L4:2785 suppresses
  2. L4:1843 weakly suppresses
  3. L5:1392 weakly excites
  4. L4:2774 weakly excites
  5. L5:1699 weakly excites
  6. L9:2261 weakly excites
  7. L16:596 weakly excites
  8. L7:738 weakly suppresses
  9. L4:666 weakly suppresses the digit 0 in dates
  10. L12:2637 weakly excites
  11. L7:1035 weakly excites
  12. L23:181 weakly suppresses
  13. L13:449 weakly suppresses
  14. L21:244 weakly suppresses
  15. L10:785 weakly excites
  16. L22:1874 weakly suppresses
  17. L7:927 weakly suppresses
  18. L3:1748 weakly suppresses "one" as a pronoun after a verb
  19. L10:818 weakly excites
  20. L18:1596 weakly suppresses
  21. L8:1096 weakly suppresses
  22. L19:751 weakly suppresses the article 'a' before a noun
  23. L11:2529 weakly excites 'here' in 'read more here' links
  24. L16:1049 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 -1.41 and -1.351 values between -0.72 and -0.652 values between -0.46 and -0.404 values between -0.40 and -0.3410 values between -0.34 and -0.2727 values between -0.27 and -0.2175 values between -0.21 and -0.15226 values between -0.15 and -0.09714 values between -0.09 and -0.021733 values between -0.02 and 0.04775 values between 0.04 and 0.11248 values between 0.11 and 0.1765 values between 0.17 and 0.2318 values between 0.23 and 0.294 values between 0.29 and 0.361 values between 0.36 and 0.421 values between 0.48 and 0.551 values between 0.55 and 0.611 values between 0.99 and 1.051 values between 1.05 and 1.12offnegativepositiveoften

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. utta
  2. utive
  3. structors
  4. utations
  5. @Before
  6. utable
  7. ☰
  8. 腾
  9. paused
  10. pa

Suppresses

  1. isel
  2. _planes
  3. jer
  4. zie
  5. ActionCreators
  6. laştır
  7. sdale
  8. axes
  9. estring
  10. sess