Neuron Explorer

Qwen3-0.6B

"that" after "such" or "such a way"

A neuron in layer 4 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 "that" when it follows "such" or "in such a way," as in "numbers such that" or "so that." It is mostly used in maths and logic sentences. Turning it up did not produce a clear theme in the model's replies.

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

    ) where they proved that a(n) is not more than the smallest number k such that the total number of ones in the binary expansion of numbers from 1 to k

  2. #2 peak

    Where is the liturgy celebrated?<|endoftext|>A particle is moving along the parabola 4y = (x + 2)^2 in such a way that its x-coordinate is increasing at a constant rate of 2 units per second.

  3. #3 peak

    77, the sequence of numbers such that their powers of 2 do not contain zeros: 1, 2,

  4. #4 peak

    holeLanguage brain teasers are those that involve the English language. You need to think about and manipulate words and letters.⏎Enter words into the following word chain such that each pair of words in the chain forms a compound word. No word can appear

  5. #5 peak

    and the operation of convolution define a commutative ring. Any ring without divisors of zero can be extended to a quotient field or field of fraction such that⏎The convolution quotient of a function with itself; i.e., f/f;

  6. #6 peak

    a Liki-design? To get an idea try to copy one of the Liki-designs on squared paper in such a way that each unit square has only one colour. What can be said about the distribution of

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: a very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very

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

    real art” only comes from the West. Introductory discussions of the history of art history and geographic conventions set the stage for considering the larger question, “where does art come from,” and other questions such as, “why haven’t I heard

  2. #2 peak

    . carbon emissions in half by 2030 and doing so equitably.⏎But neither plan explicitly grapples with the politically thorny question of where to invest. Should the money follow the people, who have flocked for decades to

  3. #3 peak

    that those drawbacks could be cured by sound engineering.⏎And so he set himself the task of creating and manufacturing a revolutionary new automobile, the Hughes Steamer.⏎Where to start?⏎Howard

  4. #4 peak

    made news when it was shut down for building code violations. Part of the center’s role was to provide accommodations for clients to “do the month.” So where do these postpartum rituals come from and are they good for new moms?⏎Researchers

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. L3:1390 excites 'that' in 'so that' purpose clauses
  2. L3:1953 excites 'such' after 'no' or 'any'
  3. L3:2501 weakly suppresses
  4. L1:2738 weakly suppresses
  5. L3:2598 weakly excites
  6. L3:866 weakly excites
  7. L3:2069 weakly suppresses
  8. L3:30 weakly excites
  9. L3:2638 weakly excites
  10. L1:1029 weakly excites
  11. L2:1603 weakly excites
  12. L3:572 weakly excites 'that' after 'so ... that'
  13. L3:1763 weakly excites
  14. L1:1024 weakly excites
  15. L3:239 weakly excites
  16. L1:1475 weakly suppresses
  17. L3:1277 weakly suppresses
  18. L1:2629 weakly suppresses
  19. L2:2692 weakly excites
  20. L1:1069 weakly suppresses
  21. L2:2929 weakly suppresses
  22. L3:2952 weakly excites
  23. L1:1830 weakly suppresses
  24. L3:2037 weakly excites

Feeds later layers

  1. L6:2466 excites
  2. L6:1199 excites
  3. L7:1673 excites
  4. L18:2056 excites
  5. L19:2437 excites
  6. L9:932 weakly excites
  7. L6:830 weakly excites
  8. L17:2225 weakly suppresses unclear: verbs after subject, line breaks
  9. L11:1470 weakly suppresses
  10. L14:1057 weakly excites
  11. L8:2635 weakly excites
  12. L21:1879 weakly excites
  13. L13:895 weakly excites
  14. L5:1064 weakly excites
  15. L18:1612 weakly excites
  16. L8:132 weakly excites
  17. L5:2727 weakly excites
  18. L15:2416 weakly excites
  19. L6:1304 weakly excites
  20. L14:2405 weakly suppresses
  21. L6:1663 weakly excites
  22. L15:504 weakly excites
  23. L7:186 weakly suppresses
  24. L17:2671 weakly excites the word matter in no matter phrases

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 -1.10 and -0.981 values between -0.86 and -0.751 values between -0.75 and -0.634 values between -0.63 and -0.513 values between -0.51 and -0.3925 values between -0.39 and -0.27206 values between -0.27 and -0.15850 values between -0.15 and -0.031745 values between -0.03 and 0.09834 values between 0.09 and 0.20188 values between 0.20 and 0.3230 values between 0.32 and 0.447 values between 0.44 and 0.564 values between 0.56 and 0.681 values between 0.68 and 0.801 values between 1.03 and 1.151 values between 1.27 and 1.391 values between 1.51 and 1.631 values between 1.63 and 1.751 values between 1.75 and 1.871 values between 2.10 and 2.221 values between 2.58 and 2.701 values between 3.53 and 3.65offnegativepositiveoften

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. positivity
  2. positive
  3. interesse
  4. positively
  5. ParameterValue
  6. Positive
  7. 有条件的
  8. Positive
  9. passed
  10. posit

Suppresses

  1. atures
  2. 씩
  3. 也不错
  4. 舐
  5. ths
  6. '),'
  7. mits
  8. تين
  9. 也将
  10. 也知道