Neuron Explorer

Qwen3-0.6B

opening parenthesis in citations

A neuron in layer 12 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 opening bracket that comes right after an author name in a reference list, such as 'Dyson, A. (2008)'. It is mostly quiet elsewhere, and the negative side shows up on words like 'this' or 'other' in technical sentences. Turning it up gave text with no clear link to citations.

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

    cornerstone of all subject areas and the Common Core. Revising to create “showing” sentences systematically helps improve writing competence.⏎Dyson, A. (2008). Playing with textual toys. In Flood, J., Heath

  2. #2 peak

    upplement 5): S213–S256.⏎- Redberg RF, et al. (2009). ACCF/AHA 2009 Performance measures

  3. #3 peak

    times of several years. The total energy radiated in a superburst is 1042 erg. Superbursts<|endoftext|>Kang, T-J. (2018) Online learning in North Korea The Diplomat, May

  4. #4 peak

    can combine this with all the other skills that are measured so that a good picture of and for the student emerges.⏎<|endoftext|>Junaidi, Juhazren (2007) Assisting learning using computers through the application of animated ped

  5. #5 peak

    of their writing. The same principle applies to adolescents and young adults.⏎- External stimuli prompt students to construct their own sentences and stories.⏎Smith, C.H. (2010). “Diving in deeper”: Bringing basic writers’ thinking

  6. #6 peak

    ⏎Lin CC, Lin XM, Zou Y, Zhang L (2005) Study on the relationship between meteorological conditions and acid rain

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 think I have been thinking for 10 minutes, and I have been thinking... and I have been thinking... and... 10 minutes, and I have been thinking... and... 10 minutes... and... 1

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

    was the most important covariate determining survival of adult D. couperi, suggesting lower survival for larger individuals within each sex. We are uncertain of the mechanisms underlying this result, but possibilities may include greater resource needs for larger individuals within each sex,

  2. #2 peak

    conversion speed regulation of the AC asynchronous motor, improve the operation precision, change the power factor, and Realize overcurrent, overvoltage, overload protection and other functions. The main circuit of the frequency converter can be roughly divided into two categories

  3. #3 peak

    classified as passive or active. In principle, a passive tracer blindly follows the fluid phase in which it is injected. Interpretation of tracer-production curves must account for this. The results from the application of active tracers may give information about fluid saturation

  4. #4 peak

    colleagues demonstrated that the tool could reveal signs of pulmonary fibrosis in COVID long-haulers, thus helping to account for shortness of breath, coughing and other lung troubles. A diagnosis, he suggests, that would be impossible with standard CT

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. L9:1059 weakly suppresses
  2. L6:1629 weakly excites
  3. L2:1203 weakly suppresses
  4. L11:272 weakly excites
  5. L10:1202 weakly excites
  6. L6:1282 weakly suppresses
  7. L11:2912 weakly suppresses comma closing an introductory clause
  8. L6:2654 weakly suppresses
  9. L5:2497 weakly excites
  10. L8:303 weakly suppresses
  11. L10:2913 weakly excites
  12. L3:1068 weakly excites "that" pointing back to a noun
  13. L10:1542 weakly suppresses
  14. L0:1983 weakly suppresses
  15. L10:2144 weakly excites
  16. L4:2470 weakly excites
  17. L1:2360 weakly excites
  18. L6:1974 weakly excites
  19. L6:454 weakly suppresses
  20. L9:1878 weakly suppresses
  21. L7:504 weakly excites
  22. L1:1090 weakly excites
  23. L5:2689 weakly suppresses
  24. L8:1219 weakly suppresses

Feeds later layers

  1. L14:1402 suppresses
  2. L24:150 suppresses
  3. L16:284 suppresses
  4. L17:1638 suppresses
  5. L16:2165 suppresses
  6. L17:1858 suppresses
  7. L16:1711 suppresses
  8. L13:641 suppresses
  9. L18:156 suppresses
  10. L21:124 excites
  11. L19:1616 weakly suppresses
  12. L17:1022 weakly suppresses
  13. L18:675 weakly suppresses
  14. L16:675 weakly suppresses
  15. L14:2349 weakly excites
  16. L21:1722 weakly suppresses
  17. L20:1372 weakly suppresses
  18. L15:1437 weakly suppresses
  19. L13:1320 weakly suppresses
  20. L14:2156 weakly suppresses opening parenthesis after a term
  21. L25:339 weakly suppresses
  22. L20:271 weakly excites
  23. L14:825 weakly suppresses
  24. L18:2173 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 -2.29 and -2.212 values between -1.56 and -1.471 values between -1.39 and -1.311 values between -1.23 and -1.152 values between -1.15 and -1.071 values between -1.07 and -0.991 values between -0.99 and -0.905 values between -0.90 and -0.822 values between -0.82 and -0.741 values between -0.74 and -0.662 values between -0.66 and -0.583 values between -0.58 and -0.507 values between -0.50 and -0.4236 values between -0.42 and -0.3387 values between -0.33 and -0.25184 values between -0.25 and -0.17422 values between -0.17 and -0.09744 values between -0.09 and -0.011025 values between -0.01 and 0.07680 values between 0.07 and 0.16392 values between 0.16 and 0.24190 values between 0.24 and 0.3268 values between 0.32 and 0.4039 values between 0.40 and 0.486 values between 0.48 and 0.561 values between 0.56 and 0.642 values between 0.64 and 0.731 values between 0.81 and 0.892 values between 0.89 and 0.97offnegativepositiveoften

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. WithEmail
  2. istory
  3. �
  4. Ballet
  5. izard
  6. glColor
  7. 甲状
  8. fä
  9. ogn
  10. 遇上

Suppresses

  1. systemic
  2. empre
  3. hai
  4. riors
  5. REW
  6. arching
  7. finity
  8. ALSE
  9. lesi
  10. corporate