Neuron Explorer

Qwen3-0.6B

'without' introducing a contrast

A neuron in layer 17 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 word 'without' when it links two ideas, like 'reduce the temperature without generating steam'. When it was turned up, the model kept repeating words like 'being', so the effect is loose.

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

    to balance this equaliser? You can put more⏎than one weight on a hook.⏎What is the greatest number of counters you can place on the grid below without four of them lying at the corners of a square?⏎How many different triangles can

  2. #2 peak

    marked transformation with the discovery of an aspect of enzyme evolution called "moonlighting." That is, a protein can acquire a second (moonlighting) function without concomitantly losing all or part of its original function

  3. #3 peak

    –though not excessive–application of inorganic fertilizer boosted rice yields and reduced land use and water use, all without substantially increasing greenhouse gas emissions. The findings also showed that high-yield farming often

  4. #4 peak

    , a quick two-to-three-second dash with a straight stream will reduce the temperature without generating a massive amount of steam.3⏎The direct attack method is designed to

  5. #5 peak

    values in the IO or (ST s) monads, if you need them.⏎> That is, I can't change a⏎> component of somethingNew without creating y. The point is there's⏎> nothing about x that changed,⏎

  6. #6 peak

    in this woman or her story. But then she gets to Africa and everything shifts. Caitlin must find a way to keep elephants out of the farmers’ crops without

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: the number of, as a whole, of, and, and, and, and, and, and, and, and, and, and, and, and, and, and, and, and, and, and, and,

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

    , historical analysis, and any understanding of broader systemic relations. In this regard, it does the opposite of critical memory work by eliminating those public spheres where people learn to translate private troubles into public issues. That

  2. #2 peak

    knows more about his or her learning strategy than anyone else. So if you'd like to learn something about the best strategy to use when teaching your students to fly, why not just ask them to tell you?⏎The next time you're

  3. #3 peak

    that changes how the⏎cd command works. Normally when you try and cd⏎into a directory, it will only try and change to the path specified

  4. #4 peak

    the free Adobe Acrobat Reader.)⏎Click for the answers to this week's puzzle.⏎Click for Sudoku Puzzles from previous weeks.⏎Click for tips for teaching students to do Sudoku puzzles.⏎Copyright © 2006 Education World<|endoftext|>1.Model

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. L16:797 suppresses
  2. L6:2882 suppresses
  3. L13:2681 excites
  4. L16:1153 suppresses
  5. L16:513 suppresses
  6. L6:2064 suppresses
  7. L6:2747 suppresses
  8. L14:526 excites
  9. L6:1133 excites "to" before an infinitive verb
  10. L1:943 weakly suppresses
  11. L9:226 weakly suppresses 'to' starting an infinitive
  12. L15:136 weakly suppresses
  13. L10:707 weakly excites
  14. L15:210 weakly excites
  15. L14:1091 weakly excites
  16. L15:1186 weakly excites
  17. L15:27 weakly excites
  18. L11:223 weakly suppresses
  19. L16:314 weakly excites
  20. L13:44 weakly excites inside web addresses and citations
  21. L11:1300 weakly excites
  22. L5:763 weakly suppresses
  23. L6:2407 weakly suppresses
  24. L8:1703 weakly suppresses

Feeds later layers

  1. L24:140 strongly excites
  2. L22:1425 strongly excites
  3. L22:609 strongly excites
  4. L26:686 strongly excites
  5. L21:1441 strongly excites
  6. L23:424 excites
  7. L25:620 excites
  8. L27:410 excites
  9. L25:376 excites
  10. L27:1237 excites
  11. L26:252 excites
  12. L27:1890 excites
  13. L25:218 excites
  14. L27:677 excites
  15. L26:708 excites
  16. L27:1148 excites
  17. L21:1521 suppresses
  18. L26:808 excites
  19. L27:1545 excites
  20. L27:303 excites
  21. L25:1596 excites
  22. L26:1304 excites
  23. L24:1811 excites
  24. L26:733 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 -7.50 and -6.912 values between -5.74 and -5.161 values between -5.16 and -4.574 values between -4.57 and -3.9911 values between -3.99 and -3.4026 values between -3.40 and -2.8236 values between -2.82 and -2.2382 values between -2.23 and -1.65129 values between -1.65 and -1.07260 values between -1.07 and -0.482017 values between -0.48 and 0.111127 values between 0.11 and 0.6970 values between 0.69 and 1.2745 values between 1.27 and 1.8633 values between 1.86 and 2.4415 values between 2.44 and 3.039 values between 3.03 and 3.616 values between 3.61 and 4.204 values between 4.20 and 4.786 values between 4.78 and 5.376 values between 5.37 and 5.952 values between 5.95 and 6.545 values between 7.12 and 7.711 values between 7.71 and 8.291 values between 9.46 and 10.041 values between 10.04 and 10.632 values between 10.63 and 11.211 values between 11.80 and 12.381 values between 12.38 and 12.971 values between 12.97 and 13.551 values between 14.72 and 15.312 values between 15.31 and 15.89offnegativepositiveoften

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. being
  2. needing
  3. having
  4. being
  5. requiring
  6. providing
  7. having
  8. étant
  9. allowing
  10. preferring

Suppresses

  1. ichten
  2. iless
  3. 我会
  4. oster
  5. arose
  6. "]];⏎
  7. ']];⏎
  8. by
  9. saja
  10. veillance