Neuron Explorer

Qwen3-0.6B

'to' after 'not' (not to do)

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 fires on the word 'to' when it comes right after 'not', as in 'not to disturb' or 'not to go out'. Turning it up did not produce clearly related writing; the outputs wandered off in different directions.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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

    lightly so as not to disturb the soil. Continue adding water until the excess flows from the bottom drainage holes

  2. #2 peak

    which helps you lose weight. If calories consumed in food are less than calories spent during exercise and other daily activities, you will lose weight. It is impossible not to! Many people have learned how to lose belly fat by starting a running program.⏎

  3. #3 peak

    in the country -- in that respect, it’s a pretty average state.⏎Largest Cities in West Virginia⏎West Virginia is one of the few US states not to have a city with more than 100,000 residents.

  4. #4 peak

    so that she is certain not to go out of the lines while another little girl with the same princess coloring pages scri

  5. #5 peak

    regulations for businesses. Its leader, deputy leader and finance spokesperson, Tim Farron, enjoys a strong base of support in the<|endoftext|>Many of us are inclined not to talk about things that upset us. We try to put a lid on our feelings

  6. #6 peak

    past”⏎“Look forward, not back”⏎All said by people who seemingly think that experiences of trauma can be consciously resolved, if we just try hard enough not to think about them, as if it were a choice to be impacted by our worst

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: I have a 4-day event. The first day, I do a task that involves a game, and the second day also, I have a different, more complex task. The goal is to get a full salary. However, 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

    together⏎- Singing songs⏎- Nursery Rhymes⏎- Question and answer board games⏎- Te Reo Singalong with Sharon Holt⏎- Listent to stories – a collection of free children’s audio books. All written and produced in

  2. #2 peak

    make less than the top 25 percent of high school graduates.⏎This alternative reading of the figure helps to explain why many young people have not responded so enthusiastically to the “market signals” about college. On average, college pays off, but

  3. #3 peak

    to produce a treatment response as olanzapine. Both positive symptoms (psychosis) and negative symptoms (blunted emotional response, reduced motivation) responded better to clozapine. John H. Krystal, M.D., Editor of Biological

  4. #4 peak

    . Curiously, it was the king of Egypt, and not God, or the Israelites themselves, who coined the phrase, "Israelite People." Acting out of panic, Pharaoh wanted to do away with the Hebrews, when he

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. L3:1866 weakly suppresses
  2. L3:122 weakly excites
  3. L3:59 weakly excites
  4. L3:421 weakly suppresses
  5. L3:370 weakly suppresses
  6. L1:1507 weakly excites
  7. L3:456 weakly suppresses
  8. L0:1372 weakly excites
  9. L1:2438 weakly excites
  10. L0:2721 weakly excites
  11. L3:115 weakly suppresses
  12. L3:272 weakly excites
  13. L2:1275 weakly excites
  14. L3:91 weakly excites
  15. L0:966 weakly suppresses
  16. L1:1973 weakly suppresses
  17. L1:318 weakly excites
  18. L0:2726 weakly excites
  19. L2:1708 weakly suppresses
  20. L3:2212 weakly excites
  21. L0:2280 weakly suppresses
  22. L3:71 weakly excites
  23. L1:1051 weakly suppresses
  24. L0:1572 weakly excites

Feeds later layers

  1. L18:2419 weakly excites
  2. L15:1622 weakly suppresses
  3. L23:217 weakly suppresses
  4. L18:141 weakly suppresses
  5. L8:723 weakly excites
  6. L18:2755 weakly suppresses
  7. L5:1958 weakly excites
  8. L26:836 weakly suppresses
  9. L15:2198 weakly excites
  10. L17:187 weakly excites
  11. L17:2097 weakly excites
  12. L15:397 weakly excites
  13. L13:1892 weakly suppresses
  14. L9:704 weakly suppresses the zero completing 10 or 100
  15. L6:439 weakly excites
  16. L24:1093 weakly excites
  17. L7:681 weakly excites 'from' after prevent or stop
  18. L14:2455 weakly excites
  19. L17:2466 weakly excites
  20. L24:2863 weakly excites
  21. L5:53 weakly excites
  22. L20:1652 weakly excites
  23. L20:530 weakly suppresses
  24. L6:3070 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.

2 values between -1.55 and -1.411 values between -1.28 and -1.151 values between -1.15 and -1.022 values between -1.02 and -0.882 values between -0.88 and -0.752 values between -0.75 and -0.6212 values between -0.62 and -0.4937 values between -0.49 and -0.35149 values between -0.35 and -0.22668 values between -0.22 and -0.091982 values between -0.09 and 0.04793 values between 0.04 and 0.18157 values between 0.18 and 0.3138 values between 0.31 and 0.4419 values between 0.44 and 0.579 values between 0.57 and 0.715 values between 0.71 and 0.846 values between 0.84 and 0.973 values between 0.97 and 1.104 values between 1.10 and 1.242 values between 1.24 and 1.377 values between 1.37 and 1.501 values between 1.50 and 1.633 values between 1.63 and 1.771 values between 1.90 and 2.031 values between 2.30 and 2.431 values between 3.62 and 3.75offnegativepositiveoften

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. .tf
  2. "#
  3. ук
  4. 到时候
  5. shade
  6. 选用
  7. 去
  8. nữa
  9. .Xr
  10. tensorflow

Suppresses

  1. otive
  2. loyment
  3. entire
  4. oting
  5. ritch
  6. alling
  7. aly
  8. antage
  9. oly
  10. abble