Neuron Explorer

Qwen3-0.6B

'much' or 'many' after 'not' or 'little'

A neuron in layer 7 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 responds to words like 'much', 'many' and 'little' when they follow a 'not' or a 'have not changed' style phrase, meaning a small amount. Turning it up did not produce text that clearly shows this idea.

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

    52, if you wanted to create a list of users, you would create a new variable and:⏎Now with the List data type, things haven’t changed much. To create a List of Users, you create a new variable and:⏎The

  2. #2 peak

    add another 21.5 cents per gallon. In addition, many local governments levy additional sales and other taxes on gasoline. Gasoline taxes have not risen much as the price of gasoline exploded upward.⏎The price of gasoline is much lower than

  3. #3 peak

    one hope beyond the grave; this was God’s promise to Abraham. God gave Moses the Law while Israel wandered in the desert. It was rigid. There was little flexibility in its demands. It declared “an eye for an eye and a tooth

  4. #4 peak

    1941, the first graduation of doctors of our institute was held, young specialists immediately after graduation went to the front to save lives. Unfortunately, not many of them managed to return.”⏎The participants of the studio “Door to

  5. #5 peak

    , not supposed to have a head. Ordinary people are just bones. The spear is the novelty, nevertheless still the gestures are ritual symbolism.⏎We do not know much about the rule of the Romans in the provinces. However we know that they in

  6. #6 peak

    Digital Download 7 Day Vegan Meal Plan | Digital download⏎Awesome Benefits Of The Paleo Diet⏎The idea behind the paleo diet is that we haven’t changed much since the time of the earliest humans. It is said that “From a genetic

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 13. It is a true statement. The question is whether such a thing as a "thing else" exists. So, for example, if the same location, time, and date, and only if the same as

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

    of the container. This usually occurs in about 1-2 hours depending on the size of your jar. A full melt

  2. #2 peak

    ans' donations.⏎The eagles typically incubate their eggs for about 35 days, and this year on the EagleCam website the DNR

  3. #3 peak

    Bay roughly 30 percent of the setnet permit holders are women. NOAA Fisheries scientists from

  4. #4 peak

    the Study of Religion" by Dr. Clayton Crockett. In it<|endoftext|>Approximately 1.6 million people die worldwide every year as a result of pneumococcal infection, which

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. L5:289 strongly suppresses
  2. L5:1087 suppresses
  3. L4:656 excites
  4. L4:3049 suppresses
  5. L5:3048 excites
  6. L6:1132 excites
  7. L4:1678 suppresses
  8. L6:222 suppresses
  9. L5:2545 excites the zero in round numbers like 100
  10. L6:2805 weakly excites
  11. L4:955 weakly excites
  12. L5:218 weakly suppresses
  13. L6:2646 weakly suppresses
  14. L4:1543 weakly suppresses
  15. L1:956 weakly excites
  16. L6:1116 weakly excites
  17. L6:572 weakly suppresses
  18. L1:1289 weakly suppresses
  19. L5:2787 weakly suppresses 'and' in 'more and more' pairs
  20. L6:2033 weakly excites
  21. L2:2714 weakly excites
  22. L2:2193 weakly suppresses
  23. L0:2517 weakly suppresses
  24. L5:2100 weakly excites

Feeds later layers

  1. L16:1273 strongly excites
  2. L12:1155 strongly excites
  3. L9:230 excites
  4. L8:1643 excites
  5. L14:1875 excites
  6. L10:82 suppresses
  7. L12:494 suppresses
  8. L12:1296 suppresses
  9. L9:781 excites
  10. L9:620 excites
  11. L11:1674 excites
  12. L13:2393 excites
  13. L18:623 suppresses
  14. L11:640 excites
  15. L18:940 excites
  16. L10:644 excites
  17. L11:922 excites
  18. L12:2068 excites
  19. L15:2651 suppresses
  20. L11:1106 excites
  21. L21:542 excites
  22. L19:1393 excites
  23. L8:1049 excites
  24. L16:1720 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 -5.21 and -4.991 values between -3.69 and -3.472 values between -3.26 and -3.042 values between -3.04 and -2.821 values between -2.82 and -2.612 values between -2.61 and -2.391 values between -2.39 and -2.174 values between -2.17 and -1.964 values between -1.96 and -1.749 values between -1.74 and -1.5210 values between -1.52 and -1.308 values between -1.30 and -1.0927 values between -1.09 and -0.8731 values between -0.87 and -0.6583 values between -0.65 and -0.44342 values between -0.44 and -0.221748 values between -0.22 and -0.001293 values between -0.00 and 0.22154 values between 0.22 and 0.4385 values between 0.43 and 0.6534 values between 0.65 and 0.8717 values between 0.87 and 1.0914 values between 1.09 and 1.3012 values between 1.30 and 1.526 values between 1.52 and 1.743 values between 1.74 and 1.954 values between 1.95 and 2.173 values between 2.17 and 2.392 values between 2.39 and 2.602 values between 2.60 and 2.821 values between 2.82 and 3.041 values between 3.04 and 3.261 values between 3.26 and 3.47offnegativepositiveoften

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. glm
  2. ilit
  3. /security
  4. itez
  5. spar
  6. iards
  7. ris
  8. diff
  9. security
  10. bé

Suppresses

  1. -equiv
  2. valid
  3. full
  4. _VERTICAL
  5. stalls
  6. state
  7. 奋斗目标
  8. ận
  9. PostBack
  10. fate