Neuron Explorer

Qwen3-0.6B

"is" in "what is" questions

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 fires on the word "is" when it comes right after "What" in a question, like "What is the meaning of...?" Its opposite side fires on commas after phrases like "the question is," which points to a shared theme of questions. Turning it down made the model repeat "what" or "how" over and over.

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

    There is now firm evidence supporting the lever arm hypothesis for the origin of directional movement. Nevertheless, many pivotal issues remain. For example, what is the sequence of events in the conversion of the chemical energy derived from ATP hydroly

  2. #2 peak

    poem serves as a reminder of the importance of connecting with nature, finding solace in its embrace, and seeking a deeper understanding of oneself.⏎Q1: What is the meaning behind Mary Oliver’s poems?⏎A1: Mary Oliver’s poems often

  3. #3 peak

    There⏎are two rows of seats in the waiting room. Using the rules, where⏎are they all sitting? Can you find all the possible ways?⏎What is the best way to shunt these carriages so that each train⏎can continue

  4. #4 peak

    Hittites?⏎- What is the meaning of Hittites?⏎- What does Itin mean?⏎- What

  5. #5 peak

    <|endoftext|>What is Cancer? What Causes Cancer?⏎Cancer is a class of diseases characterized by out

  6. #6 peak

    .What is the difference between MOV and MVI?⏎12. What are the functions 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: How can what what how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how how

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 there is not much chance of getting the genie back in the bottle. It is here and you better believe it’s impacting your business. The question is, what are you going to do about it?⏎The term “Green” comes from

  2. #2 peak

    , the organic acids that are part of each kefir grain readily react with reactive metals. The important question is, which metals are reactive in the process of making and storing kefir? The answer

  3. #3 peak

    to move beyond simply complaining, reflexively and vaguely, that government is "too big." They must instead develop a concrete, practical approach to answering the question, "What do you propose to cut?" — a realistic set of guidelines that elucidate

  4. #4 peak

    will open two charter schools in Baton Rouge. Like many parents, you may be wondering what exactly is a charter school? And more importantly, is it a good educational option for my child?⏎What is a Charter School?⏎

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. L17:2638 strongly suppresses
  2. L17:1480 strongly excites
  3. L16:2995 strongly excites
  4. L18:1586 suppresses
  5. L18:2010 excites
  6. L17:1681 excites
  7. L17:1407 excites "and" joining questions (wh-question context)
  8. L12:2540 suppresses
  9. L7:2761 suppresses
  10. L18:2562 excites
  11. L15:1871 excites
  12. L7:2058 excites
  13. L18:1502 suppresses
  14. L14:1800 suppresses
  15. L18:2666 suppresses
  16. L17:1952 weakly excites
  17. L16:1942 weakly suppresses
  18. L16:2097 weakly suppresses
  19. L1:1464 weakly excites
  20. L18:3053 weakly excites "how" after learn or know
  21. L13:2728 weakly suppresses
  22. L16:2204 weakly suppresses
  23. L8:2326 weakly excites
  24. L14:492 weakly suppresses

Feeds later layers

  1. L27:656 strongly suppresses
  2. L21:728 strongly suppresses
  3. L22:1414 strongly suppresses
  4. L23:594 suppresses
  5. L22:543 suppresses
  6. L22:2255 suppresses
  7. L26:1836 suppresses
  8. L26:400 suppresses
  9. L23:1089 suppresses
  10. L25:576 suppresses
  11. L26:1046 suppresses
  12. L20:1043 suppresses
  13. L24:1439 suppresses
  14. L22:1592 suppresses
  15. L25:1407 excites 'in' right after a place or time
  16. L21:360 suppresses 'such' introducing examples
  17. L26:1887 suppresses
  18. L20:1114 suppresses
  19. L25:2039 suppresses
  20. L23:197 weakly suppresses
  21. L23:2187 weakly suppresses
  22. L27:741 weakly suppresses
  23. L26:1104 weakly suppresses
  24. L20:73 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 -36.66 and -35.621 values between -24.26 and -23.231 values between -17.04 and -16.001 values between -13.94 and -12.901 values between -12.90 and -11.872 values between -11.87 and -10.841 values between -9.81 and -8.771 values between -8.77 and -7.742 values between -7.74 and -6.713 values between -6.71 and -5.682 values between -5.68 and -4.642 values between -4.64 and -3.615 values between -3.61 and -2.5813 values between -2.58 and -1.54170 values between -1.54 and -0.513580 values between -0.51 and 0.52110 values between 0.52 and 1.556 values between 1.55 and 2.593 values between 2.59 and 3.623 values between 3.62 and 4.65offnegativepositiveoften

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. ỗ
  2. vine
  3. iminal
  4. cular
  5. cgi
  6. 我认为
  7. ıyla
  8. azard
  9. ucas
  10. amment

Suppresses

  1. how
  2. how
  3. 如何
  4. 哪些
  5. 怎样
  6. what
  7. 怎么
  8. How
  9. 怎麼
  10. 哪