Neuron Explorer

Qwen3-0.6B

'and' in lists of places or names

A neuron in layer 16 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 'and' when it joins the last two items in a list, especially lists of countries, places or names. Turning it up gave text full of list-like repetition, but the link to the trigger is weak.

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

    standards will be piloted in Botswana, Egypt, Ethiopia, Kenya and Paraguay. The initiative contributes to WHO’s 13th General Program of

  2. #2 peak

    based on a priori hypotheses, to examine intraspecific differences in survival probabilities relative to individual covariates (sex, size, size standardized by sex, and overwintering location). Annual survival in 2003 and 2

  3. #3 peak

    the world. This will not only benefit biodiversity, but will also help build local resilience to the increasing impacts of climate change. Three examples from the Philippines, Indonesia and Kenya are presented here.⏎Mount Siburan Important Bird Area (IBA)

  4. #4 peak

    th and adults, and their families, who are dealing with anorexia nervosa, bulimia nervosa, avoidant/restrictive food intake disorder and binge eating disorder.⏎- Clients are referred into our program by their family doctors.⏎

  5. #5 peak

    -888-782-7937)⏎For more information on activities ongoing within EPA Region 7 (Iowa, Kansas, Missouri and Nebraska), contact:<|endoftext|>According to the analysis of the cutting process oftooth

  6. #6 peak

    plan will be recommended.⏎- If needed, youth and adults will be referred to provincial tertiary facilities at St. Paul's Hospital, B.C. Children's Hospital and Looking Glass Residence.⏎- Kelty Mental Health: Eating disorders⏎Connects you

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: For example, for a different, or, for a different, or, for a different, or, for a different, or, for, for, for, for, for, for, for, for, for, for, for

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

    a significant effect of scene condition, with average threshold speed changes of 43%, 37% and 25% for conditions 1, 2 and 3, respectively. Thus, monocular depth cues aid the detection of

  2. #2 peak

    4; p < .001 for all). Through iterative thematic analysis we developed an alliterative “7P” model spanning 2 domains: (1) Participants (Performers and Peers); and (2) Program (Part

  3. #3 peak

    example were given wood and plaster in which to design and fabricate an audio speaker set.⏎Teaching Focus: The objective of this exercise was two-fold; (1) to have students work with two contrasting design materials- one common and easily manipulated

  4. #4 peak

    regardless of whether it is a speculative narrative or a narrative of emancipation (p. 37).⏎Two expressions in the quotes above catch my attention: (a) “It may be difficult for you to envision…at one time…the grand

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. L14:1175 suppresses
  2. L14:3034 suppresses
  3. L14:3041 suppresses
  4. L13:749 suppresses
  5. L14:1174 suppresses
  6. L15:935 excites
  7. L13:2742 excites
  8. L12:2071 suppresses
  9. L15:1005 excites
  10. L14:1234 suppresses
  11. L12:2057 excites
  12. L15:137 suppresses
  13. L11:1674 suppresses
  14. L15:2322 suppresses
  15. L15:1504 excites
  16. L11:2760 suppresses
  17. L13:2022 excites
  18. L10:713 excites
  19. L14:841 excites commas inside lists
  20. L14:396 suppresses
  21. L13:1063 excites
  22. L11:467 suppresses
  23. L13:1361 suppresses
  24. L12:1530 suppresses

Feeds later layers

  1. L18:2898 suppresses
  2. L18:1693 suppresses
  3. L17:2891 suppresses
  4. L24:1087 suppresses
  5. L18:2722 suppresses
  6. L19:1414 suppresses
  7. L19:1265 suppresses
  8. L18:2446 suppresses
  9. L19:2584 suppresses
  10. L18:1327 suppresses
  11. L18:957 excites
  12. L19:492 excites
  13. L17:414 excites
  14. L19:2999 suppresses
  15. L19:1092 excites
  16. L20:1441 suppresses
  17. L18:2589 suppresses
  18. L21:2512 suppresses
  19. L19:2552 suppresses
  20. L19:1813 suppresses
  21. L23:1798 suppresses
  22. L27:291 suppresses
  23. L17:1384 weakly suppresses
  24. L17:1510 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 -3.87 and -3.671 values between -3.46 and -3.261 values between -3.26 and -3.051 values between -2.85 and -2.641 values between -2.64 and -2.442 values between -2.44 and -2.231 values between -2.23 and -2.032 values between -2.03 and -1.8210 values between -1.82 and -1.625 values between -1.62 and -1.4217 values between -1.42 and -1.2129 values between -1.21 and -1.0136 values between -1.01 and -0.8055 values between -0.80 and -0.60148 values between -0.60 and -0.39463 values between -0.39 and -0.191909 values between -0.19 and 0.02889 values between 0.02 and 0.22171 values between 0.22 and 0.4352 values between 0.43 and 0.6346 values between 0.63 and 0.8416 values between 0.84 and 1.0415 values between 1.04 and 1.259 values between 1.25 and 1.457 values between 1.45 and 1.665 values between 1.66 and 1.865 values between 1.86 and 2.072 values between 2.07 and 2.274 values between 2.27 and 2.472 values between 2.47 and 2.682 values between 2.68 and 2.881 values between 4.11 and 4.32offnegativepositiveoften

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. 三个
  2. 两个
  3. 一个是
  4. 两种
  5. 三个方面
  6. 两条
  7. 两次
  8. 三种
  9. 兩個
  10. 两点

Suppresses

  1. etc
  2. etc
  3. 等等
  4. 等
  5. 等各种
  6. 等诸多
  7. など
  8. 等地
  9. 等功能
  10. 等一批