Neuron Explorer

Qwen3-0.6B

'to' after 'come' or 'jump'

A neuron in layer 15 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 'to' in phrases like 'come to a conclusion' or 'jump to conclusions', where the sentence is about reaching a conclusion. When it was turned up, the model's text drifted toward phrases like 'come up with' and 'come to'. When turned down, the text fell into repeated apologies.

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 the time of the creation of the world. Provisionally, based on the studies published in Biblical Chronology over the past two years, I have come to the year 3932 B.C. as the year of creation,

  2. #2 peak

    importance of connections between the biological diversity of the Sky Islands and that of the Sierra Madre Occidental of Mexico. Many separate scientific analyses of the region have come to the same conclusion: the biological diversity of the individual mountain ranges is in peril if

  3. #3 peak

    933 at<|endoftext|>Bedross Der Matossian⏎One of the outstanding issues in Armenian Genocide historiography has been the inability of historians to come to a consensus regarding the causes, the aim of the perpetrators, and the process of

  4. #4 peak

    fairly far down on a tree that relies on a pyramidal shape to really look nice, then it too should probably be taken out.”⏎But don’t jump to hasty conclusions, Trinklein said. Just like you can’t unring

  5. #5 peak

    of the reason they won…other than sheer suicidal bravery and determination…was another belief that underlies the two involving women discussed here: Resurrection.⏎I came to identify this concept as Maccabean. Book Two of the Books of the

  6. #6 peak

    University Police will assess the situation for danger of imminent harm.⏎2. A team led by the Dean of Students and the Chief Diversity Officer will review the report and reach out to you to provide care, offer support, and learn more about the incident

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: it's not. It's not. I don't know what I mean. But it's not. It's not. Maybe. Maybe. Maybe. (c) -) I need to know what I mean by the "d"

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

    and need to know where is the Ganges river. I know it is in India but I need more details.⏎I would really appreciate your help with this question :-)⏎Hi Rebecca, it is a good question but difficult to answer because of the

  2. #2 peak

    completely useless :P I just felt like correcting the other code (too tired to check who it was that posted it, and I can't see just now anyway :P)⏎void genFactorial(int x, int t)⏎int total = t

  3. #3 peak

    the salt out of their bodies.⏎Again, the Green Sea Turtles are a protected species and it is VERY important to NEVER TOUCH THEM!⏎Happy Diving! :)<|endoftext|>Few deny there are major global challenges ahead — from climate change and loss of

  4. #4 peak

    here about those 'Box-Shaped Watermelons'!? Or how about a yellow center instead of pinkish-red! Seedless?⏎See you next time! :-)<|endoftext|>A Safe Hospital, Reduce Risk, Save Lives⏎The Mission of Safe Hospital

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. L6:2009 excites
  2. L10:2309 weakly suppresses
  3. L2:1205 weakly suppresses
  4. L9:846 weakly excites
  5. L13:906 weakly excites
  6. L13:2684 weakly suppresses
  7. L5:1243 weakly suppresses
  8. L2:2784 weakly suppresses
  9. L2:1346 weakly suppresses
  10. L13:2864 weakly excites
  11. L7:135 weakly excites 'of' after 'issue' in citations
  12. L0:2897 weakly excites
  13. L0:793 weakly suppresses
  14. L10:2688 weakly suppresses
  15. L14:2446 weakly excites
  16. L11:1593 weakly suppresses
  17. L12:947 weakly excites
  18. L10:1485 weakly excites
  19. L6:1693 weakly excites
  20. L11:1786 weakly suppresses
  21. L1:887 weakly suppresses 'be' after 'might', 'also' or 'only'
  22. L2:275 weakly suppresses 'of' after effectiveness or instead
  23. L13:967 weakly excites
  24. L8:2829 weakly suppresses

Feeds later layers

  1. L16:592 suppresses
  2. L18:746 weakly suppresses
  3. L16:1371 weakly excites
  4. L25:2167 weakly suppresses
  5. L16:2743 weakly excites
  6. L17:1171 weakly suppresses
  7. L18:828 weakly suppresses
  8. L24:682 weakly excites
  9. L18:2068 weakly suppresses
  10. L17:2969 weakly suppresses
  11. L18:1882 weakly suppresses
  12. L17:259 weakly suppresses
  13. L25:1600 weakly excites
  14. L19:327 weakly suppresses
  15. L22:1027 weakly excites
  16. L19:1610 weakly excites
  17. L20:1453 weakly suppresses
  18. L21:1524 weakly suppresses
  19. L19:973 weakly suppresses
  20. L18:850 weakly suppresses
  21. L21:1703 weakly suppresses
  22. L25:147 weakly excites
  23. L26:2240 weakly excites
  24. L17:1286 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 -1.90 and -1.812 values between -0.95 and -0.8610 values between -0.76 and -0.6720 values between -0.67 and -0.5757 values between -0.57 and -0.48126 values between -0.48 and -0.38264 values between -0.38 and -0.29476 values between -0.29 and -0.19695 values between -0.19 and -0.10874 values between -0.10 and -0.00717 values between -0.00 and 0.09384 values between 0.09 and 0.19180 values between 0.19 and 0.2957 values between 0.29 and 0.3829 values between 0.38 and 0.489 values between 0.48 and 0.573 values between 0.57 and 0.671 values between 0.67 and 0.761 values between 0.86 and 0.951 values between 0.95 and 1.051 values between 1.81 and 1.90offnegativepositiveoften

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. 为首的
  2. \/
  3. олн
  4. 为首
  5. "]=="
  6. iode
  7. _|
  8. stime
  9. Bail
  10. =db

Suppresses

  1. 라이
  2. OLDER
  3. ingo
  4. 嘿嘿
  5. ARIO
  6. pop
  7. Osw
  8. earned
  9. ENDER
  10. AGO