Neuron Explorer

Qwen3-0.6B

'way' in 'work their way' phrases

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 lights up on the word 'way' in phrases like 'work their way up' or 'find its way'. Its negative side points to traffic lights and roundabouts. Turning it up did not produce writing that clearly connects to these phrases.

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

    throughout the world and will often latch on to shoes or clothes, and then work their way up your clothing until they find a prime piece of exposed skin. When they find

  2. #2 peak

    in more efficiently, and professors are able to use more of their time with students individually as their office schedules are easier to manage. Many students have to work their way through college to support themselves while receiving an education. This is much easier with a

  3. #3 peak

    others for survival. They would rather work for meagre salary for now and work their way to higher salaries or income positions. Those who believe they are owed something may never

  4. #4 peak

    policy.⏎The Magic School Bus Lost in the Solar System Unit Study⏎My kids are IN LOVE with the Solar System right now. So we have been working our way thru our The Magic School Bus Lost in the Solar System Unit Study. We have

  5. #5 peak

    the image of a Native American find its way to Ghent? And how should we deal with it when placed within a modern

  6. #6 peak

    ethical in a given situation if the problem is simple and straightforward. But when a complex dilemma leaves you uncertain, there are some tools that can help you work your way to a good decision. They can’t tell you what to do, but they

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 green light, and there is no way. The green light is green. The green light is red. The green light is red. The green light is red. The green light is red. The green light is red. The green

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

    The standard Teletype paper tape uses only 8 bits.) Cities would hire private companies to translate the data into reports that traffic engineers could use to adjust traffic lights or improve roads.⏎Bill Gates and Paul Allen were high school students at Lakeside

  2. #2 peak

    around a detailed communication protocol, which we also present. We demonstrate in simulation that our new mechanism has the potential to significantly outperform current intersection control technology -- traffic lights and stop signs. Because our mechanism can emulate a traffic light or stop sign,

  3. #3 peak

    deciding on whether to turn left or right at an intersection, whether to slow down when approaching a roundabout, or to stop for a pedestrian at a pedestrian crossing. A fully automated vehicle

  4. #4 peak

    ten objects from museum collections in Greater Manchester⏎A compass used in a PoW escape, a railway station sundial and part of the world's first commercial computer: objects that tell a

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. L3:1216 suppresses
  2. L3:294 weakly excites
  3. L1:3020 weakly suppresses
  4. L0:2338 weakly suppresses
  5. L3:115 weakly suppresses
  6. L3:2133 weakly excites
  7. L3:2906 weakly excites
  8. L3:1514 weakly suppresses
  9. L3:322 weakly suppresses
  10. L3:2437 weakly suppresses
  11. L2:1161 weakly suppresses
  12. L3:187 weakly excites
  13. L1:753 weakly excites
  14. L1:2650 weakly suppresses
  15. L3:703 weakly excites
  16. L1:1826 weakly suppresses
  17. L0:1301 weakly suppresses
  18. L3:567 weakly excites
  19. L2:549 weakly suppresses
  20. L0:1224 weakly suppresses
  21. L0:178 weakly suppresses
  22. L3:1327 weakly suppresses
  23. L3:2140 weakly suppresses
  24. L3:691 weakly excites

Feeds later layers

  1. L6:2200 suppresses
  2. L6:2898 suppresses
  3. L7:1674 weakly suppresses
  4. L10:2160 weakly suppresses
  5. L22:1675 weakly excites
  6. L5:1805 weakly suppresses
  7. L5:2591 weakly excites
  8. L6:2762 weakly suppresses
  9. L5:2070 weakly suppresses
  10. L5:2145 weakly suppresses
  11. L6:1226 weakly excites
  12. L7:95 weakly suppresses
  13. L6:1881 weakly suppresses
  14. L6:1471 weakly suppresses
  15. L16:2029 weakly suppresses
  16. L18:1592 weakly suppresses
  17. L13:2536 weakly excites
  18. L10:2919 weakly excites
  19. L21:2718 weakly suppresses
  20. L11:2962 weakly excites
  21. L5:1202 weakly suppresses
  22. L5:2617 weakly suppresses
  23. L15:393 weakly suppresses
  24. L7:2255 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 -3.27 and -3.132 values between -1.34 and -1.201 values between -1.07 and -0.932 values between -0.65 and -0.5213 values between -0.52 and -0.3892 values between -0.38 and -0.24538 values between -0.24 and -0.101719 values between -0.10 and 0.041179 values between 0.04 and 0.17292 values between 0.17 and 0.3146 values between 0.31 and 0.456 values between 0.45 and 0.597 values between 0.59 and 0.724 values between 0.72 and 0.862 values between 0.86 and 1.002 values between 1.14 and 1.271 values between 1.96 and 2.101 values between 2.10 and 2.24offnegativepositiveoften

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. 茵
  2. itemName
  3. 孢
  4. anded
  5. 钽
  6. 浸泡
  7. WithEmail
  8. 与中国
  9. .xmlbeans
  10. entrant

Suppresses

  1. 心动
  2. 闸
  3. ngừng
  4. kenn
  5. 買う
  6. 見る
  7. zen
  8. 례
  9. bie
  10. compr