Neuron Explorer

Qwen3-0.6B

end of document, and 'it turned out'

A neuron in layer 19 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 mostly reacts to the marker that signals the end of one document, and it also reacts weakly to the phrase 'it turned out'. Its steered outputs were unrelated and broken, so the effect of turning it up or down is hard to read.

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

    .⏎ASJC Scopus subject areas⏎- Electronic, Optical and Magnetic Materials⏎- Ceramics and Composites⏎- Condensed Matter Physics⏎- Materials Chemistry<|endoftext|>Humans are intimately connected with the physical environment, even though we have only been present

  2. #2 peak

    if you need to, you can always buy college assignments online.<|endoftext|>Breaking research news from a team of scientists led by Carnegie's Ho-kwang "

  3. #3 peak

    grammatical functions and features of Chinese auxiliary verbs.<|endoftext|>Obviously, the unlimited liability feature of general partnerships is a serious impediment to conducting

  4. #4 peak

    and ultimately foster innovation.<|endoftext|>Resource Center: Allergy Shots (Immunotherapy)⏎For people suffering severe allergic

  5. #5 peak

    Issues⏎Being able to troubleshoot common HVAC issues can be beneficial<|endoftext|>Whether intentional or not, if this process continues it will complete, perhaps finally,

  6. #6 peak

    support your existing curriculum, depending on your needs. There are cross-curricular links and activities suggested throughout.<|endoftext|>Usually, consumers don’t have problems with online financial transactions. However, occasionally something

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, I know, the number of, the number of, the number of, the number of, the number of, the number of, the number of, the number of, the number of, the number of, the

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

    produce structures that are more efficient to run, cutting ongoing costs and minimizing the carbon footprint.⏎What does this have to do with BrightBuilt Barn?⏎Well, it turns out that the Barn was also designed with (simpler and much less expensive

  2. #2 peak

    to know how EGT1 helps roots to overcome gravity. For this, the researchers looked at the genes disrupted in the EGT1 mutant. This, it turned out, were genes for producing parts of the cell wall. This disruption translated into

  3. #3 peak

    the work of a small team at IBM in Switzerland that had developed a complex technique to image or “take a picture” of molecules in the journal Nature — it turned out to be the coup de grace for the team.⏎Back in 20

  4. #4 peak

    searched for a mutant in barley that didn’t do this. And successfully so. They found a plant whose roots grew more downwards than sideways. This mutant, it turned out, had a mutation in the gene EGT1.⏎EGT1 organ

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. L18:824 suppresses
  2. L18:30 excites
  3. L17:2988 weakly excites
  4. L5:2896 weakly suppresses
  5. L2:94 weakly suppresses
  6. L18:429 weakly suppresses
  7. L15:376 weakly suppresses
  8. L18:20 weakly suppresses
  9. L6:2960 weakly suppresses
  10. L0:2926 weakly excites
  11. L2:2016 weakly suppresses
  12. L18:569 weakly excites
  13. L15:2776 weakly excites
  14. L1:2683 weakly excites
  15. L17:271 weakly excites
  16. L13:197 weakly excites
  17. L2:2621 weakly excites
  18. L8:59 weakly suppresses
  19. L9:837 weakly excites
  20. L10:65 weakly excites
  21. L7:1335 weakly suppresses
  22. L0:2329 weakly excites
  23. L4:2620 weakly excites
  24. L1:1464 weakly excites

Feeds later layers

  1. L23:236 strongly excites
  2. L26:122 strongly excites
  3. L20:841 strongly excites
  4. L27:1266 excites
  5. L25:4 suppresses
  6. L24:1896 excites
  7. L25:745 excites
  8. L25:430 excites
  9. L20:1313 excites
  10. L27:574 excites 'Aut' starts and colon prediction
  11. L24:1466 suppresses
  12. L26:1229 excites
  13. L26:1329 excites
  14. L27:646 excites
  15. L27:1695 excites
  16. L25:440 excites
  17. L20:1077 excites
  18. L27:858 weakly excites
  19. L26:648 weakly suppresses
  20. L22:940 weakly suppresses
  21. L24:563 weakly suppresses
  22. L23:275 weakly suppresses
  23. L26:485 weakly excites
  24. L23:587 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 -7.20 and -6.871 values between -5.55 and -5.231 values between -3.91 and -3.581 values between -3.58 and -3.251 values between -2.59 and -2.261 values between -1.28 and -0.9510 values between -0.95 and -0.6269 values between -0.62 and -0.292597 values between -0.29 and 0.041138 values between 0.04 and 0.3766 values between 0.37 and 0.7016 values between 0.70 and 1.031 values between 1.03 and 1.361 values between 1.36 and 1.681 values between 1.68 and 2.011 values between 3.33 and 3.661 values between 3.99 and 4.321 values between 5.63 and 5.96offnegativepositiveoften

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. ":
  2. :
  3. ':
  4. ”:
  5. ###
  6. ():
  7. ):
  8. ’:
  9. **:
  10. \":

Suppresses

  1. andel
  2. 不了
  3. ?"
  4. inho
  5. inne
  6. #$
  7. #"
  8. rones
  9. вис
  10. atel