Neuron Explorer

Qwen3-0.6B

'is' after focus or goal nouns

A neuron in layer 2 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 follows a noun like 'focus', 'goal' or 'main challenge', as in 'The focus is to...'. Turning it up did not produce a clear theme in the model's replies.

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

    skills and a “family” of supports and resources.⏎c. Inclusive ECE Coaching is intended for EHS/HS program coaches and supervisors. The focus is to build a Community of Reflective

  2. #2 peak

    Surgery. His focus is to safely return athletes to the playing field of their choice and to keep them there

  3. #3 peak

    ey. I am an experienced programmer, having received several accommodations on programming in my formal education as well as tough several programming languages to thousands of students. My goal is

  4. #4 peak

    ones as the child advances additional skills can be introduced. The focus is first set on the ability of the child to perform simple skills such as eating properly

  5. #5 peak

    1986. Now he travels around Namibia, providing stargazing and astronomy education through his company, Space Observation Learning Namibia. His goal is to have more young heads in the clouds.⏎"Nobody looks up anymore," John

  6. #6 peak

    classes to see how the students produced out with the projects. The website gives a nice collection of classroom activities that incorporate reading, writing, and math. Their focus is to make certain that all schools are living up to

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 been thinking about lately: I have I have 4 I have 5 I have 6 I have 7 8 9 1 1 2 1 2 3 4

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

    to linguistic errors are influenced by both the effectiveness and efficiency of domain-general conflict monitoring. These relationships were

  2. #2 peak

    properties. This information is especially important when enhanced-oil-recovery techniques that use expensive fluids such as surfactants, micellar fluids, or polymers are considered. In the last 50 years, many tracer studies have been reported

  3. #3 peak

    in these schools are assumed to be influenced by religion in their development, including ways of being citizens and

  4. #4 peak

    riting about those points to see which might make a useful topic. It allows them some room to explore potential topics before having to commit to one, and thus alleviates some pressure

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. L1:1100 suppresses
  2. L0:2283 excites
  3. L1:174 suppresses
  4. L1:268 excites
  5. L1:462 excites
  6. L0:406 excites
  7. L1:242 suppresses
  8. L0:632 weakly excites
  9. L1:700 weakly suppresses
  10. L1:993 weakly excites
  11. L1:505 weakly excites
  12. L1:1360 weakly excites
  13. L1:1007 weakly excites name pieces inside a multi-part word
  14. L0:670 weakly excites
  15. L1:105 weakly excites
  16. L1:579 weakly suppresses
  17. L1:239 weakly excites
  18. L1:1001 weakly excites
  19. L1:736 weakly suppresses
  20. L1:175 weakly excites
  21. L0:816 weakly suppresses
  22. L0:847 weakly suppresses
  23. L1:1579 weakly suppresses 'Well' starting a reply or answer
  24. L0:444 weakly excites

Feeds later layers

  1. L4:22 strongly suppresses
  2. L4:476 strongly excites
  3. L5:395 strongly suppresses
  4. L6:155 strongly suppresses
  5. L5:99 strongly excites
  6. L11:344 strongly excites
  7. L8:511 strongly suppresses
  8. L15:304 strongly excites
  9. L15:318 strongly suppresses
  10. L10:422 excites
  11. L5:351 suppresses
  12. L7:84 excites
  13. L9:157 suppresses
  14. L6:769 suppresses
  15. L10:586 suppresses 'important' and degree words after 'how'
  16. L16:968 excites
  17. L11:782 excites
  18. L16:881 excites
  19. L4:152 suppresses
  20. L14:850 suppresses
  21. L17:2216 suppresses
  22. L5:460 suppresses
  23. L8:1149 suppresses
  24. L10:766 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 -1.74 and -1.652 values between -1.47 and -1.391 values between -1.39 and -1.303 values between -1.30 and -1.211 values between -1.21 and -1.124 values between -1.12 and -1.035 values between -1.03 and -0.956 values between -0.95 and -0.8614 values between -0.86 and -0.7712 values between -0.77 and -0.6826 values between -0.68 and -0.6034 values between -0.60 and -0.5155 values between -0.51 and -0.42110 values between -0.42 and -0.33196 values between -0.33 and -0.25346 values between -0.25 and -0.16558 values between -0.16 and -0.071426 values between -0.07 and 0.02690 values between 0.02 and 0.11226 values between 0.11 and 0.19101 values between 0.19 and 0.2840 values between 0.28 and 0.3713 values between 0.37 and 0.4614 values between 0.46 and 0.5511 values between 0.55 and 0.637 values between 0.63 and 0.721 values between 0.72 and 0.811 values between 0.81 and 0.901 values between 0.98 and 1.071 values between 1.16 and 1.251 values between 1.25 and 1.341 values between 1.69 and 1.77offnegativepositiveoften

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. /her
  2. gì
  3. disguise
  4. &q
  5. ;z
  6. %;
  7. Dİ
  8. 一听
  9. DEFINE
  10. <nav

Suppresses

  1. 也是如此
  2. 是有
  3. antan
  4. burgh
  5. (){}⏎
  6. 中有
  7. ']),⏎
  8. }()⏎
  9. 拥
  10. るのは