Neuron Explorer

Qwen3-0.6B

'more' or 'not' before a likelihood claim

A neuron in layer 20 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on words like 'more', 'not' and 'likely' when the sentence is about one thing being more or less likely to happen, such as being more prone to injury. Turned up, the model drifted into repeating 'will' and 'going to', which fits its future-tense pull.

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

    the nerve input and how long it is lasting for a particular contraction. It is used to assess whether or not the muscle is spastic, since weak muscles are more prone to injury and spasm. Bulbocavernosus reflex latency is

  2. #2 peak

    ’s mental state. Cognitive empathy is the ability to understand that mental state. These components are independent of each other, which means a person capable of emotional empathy isn’t necessarily capable of cognitive empathy, and vice versa.⏎What Are the Issues Concerning

  3. #3 peak

    scary, but it’s not. If you’re off a few (or 30) seconds or your water’s a couple degrees off temperature, you’re NOT going to ruin your film. The more accurate you are,

  4. #4 peak

    metabolism slows down, allowing it to conserve oxygen and stay underwater longer. Similarly, the activity level of the turtle can affect its breath-holding capabilities. A resting turtle can hold its breath for longer than an active one.⏎Do turtles need to breathe

  5. #5 peak

    at tasks that they are fully able to handle successfully.⏎* Those individuals who have developed learned helplessness are far more likely to suffer from depression.⏎* Such people will tend to feel unsatisfied in

  6. #6 peak

    ⏎There are useful variants too (shown here without redirection)⏎Note: even if svn can make the difference of another directory (svn diff mydirectory), it is not recommended to do<|endoftext|>by Clinton Wee⏎Being one of the only atheist countries

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: there are some 2D that are not the best in the world, but they have the same or better features as the best 1D. What is the best 2D that we have that we are supposed to have that we

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

    <|endoftext|>The nail matrix is the layer of keratin-producing cells at the base of each toe or finger. As new cells arise, the hard husks of older cells are pushed forward, creating the nail plate – what we think of as just the

  2. #2 peak

    s and apply it to the male barnyard fowl. Slang doesn’t usually have a euphemistic quality. When we think of euphemisms, we think of words that are substituted because their connotations are less distressing than the

  3. #3 peak

    transmission of DNA via mobile genetic element is quite different from the well-known means by which parents pass on their genetic material to their offspring. In animals, pieces of DNA typically cannot jump from one adult organism's genetic material to another adult organism's genetic

  4. #4 peak

    means looking not only at political options but also at the way we humans are set up — how we’re wired. When public policies don’t account for that, we have reduced horizons, diminished resources, and polarization.⏎How are we wired?

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. L19:2133 strongly excites
  2. L17:1118 suppresses mid-clause words in technical prose
  3. L19:1862 suppresses
  4. L14:2571 excites
  5. L17:730 excites
  6. L19:553 suppresses
  7. L15:1829 excites
  8. L19:545 suppresses
  9. L19:2157 suppresses
  10. L15:1113 suppresses
  11. L19:564 suppresses
  12. L12:1702 excites
  13. L17:1356 excites
  14. L14:1347 weakly suppresses
  15. L8:2534 weakly suppresses
  16. L16:1223 weakly excites
  17. L18:233 weakly excites
  18. L10:197 weakly suppresses
  19. L6:1436 weakly suppresses "from" after a change in amount
  20. L9:281 weakly suppresses
  21. L2:270 weakly suppresses
  22. L12:385 weakly excites
  23. L19:2475 weakly suppresses
  24. L15:1513 weakly suppresses

Feeds later layers

  1. L23:765 strongly excites
  2. L27:1015 strongly excites
  3. L25:2368 strongly excites
  4. L23:1555 strongly excites
  5. L22:745 strongly suppresses
  6. L23:1173 excites
  7. L26:2680 excites
  8. L24:1106 excites
  9. L22:2824 excites
  10. L23:103 suppresses
  11. L21:1699 excites
  12. L22:630 excites
  13. L21:2523 excites
  14. L22:2166 excites
  15. L25:2151 excites
  16. L21:2121 excites
  17. L23:2305 excites
  18. L23:1419 excites
  19. L24:277 suppresses
  20. L22:2224 weakly excites
  21. L22:2399 weakly excites
  22. L26:1844 weakly suppresses
  23. L26:2147 weakly suppresses
  24. L22:1616 weakly 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 -13.31 and -12.911 values between -10.53 and -10.141 values between -9.74 and -9.342 values between -7.76 and -7.362 values between -7.36 and -6.973 values between -6.97 and -6.574 values between -6.57 and -6.182 values between -6.18 and -5.784 values between -5.78 and -5.386 values between -5.38 and -4.995 values between -4.99 and -4.599 values between -4.59 and -4.203 values between -4.20 and -3.809 values between -3.80 and -3.404 values between -3.40 and -3.0115 values between -3.01 and -2.6110 values between -2.61 and -2.2119 values between -2.21 and -1.8234 values between -1.82 and -1.4237 values between -1.42 and -1.0377 values between -1.03 and -0.63306 values between -0.63 and -0.231819 values between -0.23 and 0.161276 values between 0.16 and 0.56227 values between 0.56 and 0.9623 values between 0.96 and 1.353 values between 1.35 and 1.755 values between 1.75 and 2.141 values between 2.14 and 2.54offnegativepositiveoften

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. will
  2. will
  3. 将会
  4. Will
  5. ’ll
  6. Will
  7. 将在
  8. 'll
  9. 将
  10. WILL

Suppresses

  1. reon
  2. ают
  3. ])]⏎
  4. 炳
  5. ennent
  6. ]){⏎
  7. ائم
  8. jq
  9. doesn
  10. ious