Neuron Explorer

Qwen3-0.6B

opening parenthesis after a phrase

A neuron in layer 25 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 an opening bracket '(' that comes right after a word or phrase, usually to add an explanation. Turned up, the model produced a long string of the letter 'i'. Turned down, it wrote ordinary sentences.

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

    “REFUTABILITY”.⏎While scientists are formulating their respective hypotheses, the only way to be objective is to to find out if the hypothesis is falsifiable (

  2. #2 peak

    threaded gauge will inspect a threaded part based mainly on the pitch diameter. As long as a GO gauge is answered (threads properly without being forced) and a NO-GO gauge is not answered,

  3. #3 peak

    -level logic circuit that implements a sum of products as shown in Figure 5.11 (a). We may redraw any two-level circuit using a regular structure ( Figure 5.11 b): a vector of buffers, followed by a

  4. #4 peak

    current.⏎Semiconductors are the individual "on-off" switches that make up a electronic devices. By only allowing the conductance of electricity in specific situations (as opposed to conductors and insulators), semiconductors can produce complex

  5. #5 peak

    term Kabinett was officially noted in German wine law, and it was given its current definition which applies to wines which are light and non-chaptalized – chaptalization is the process of adding sugar to unfermented grape must

  6. #6 peak

    protection.It’s simple: the higher the number, the more protection you get. UPF50, for example, means that only 1/50 (2%) of UV Rays will pass through the fabric. As a general rule of

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: what is the relationship between the origin of the origin and the origin of the origin? Also, what is the origin of the origin, and what is the origin of the origin origin? We have already learned that the origin of the origin is

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

    almonds instead of unhealthy foods and lose weight by eating the tasty treat.⏎Another of the health benefits of almonds are good for brain development. The content of tryptoph

  2. #2 peak

    versal of dysbiosis as a therapeutic model in MS⏎- Roles for short chain fatty acids, bile acids, phytoestrogens, tryptophan, choline and mucin degradation⏎And for those readers interested in data

  3. #3 peak

    minbar, except in the case of istisqaa’ (praying for rain). And the same applies when making du’aa’ when doing tawaaf or sa’ee.⏎When there is evidence to show that it is permissible

  4. #4 peak

    's syndrome, the use of haloperidol is limited to the most serious cases.⏎Other medications in this group such as olanzapine (Zyprexa), risperidone (Risperdal), ziprasidone (

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. L19:158 suppresses
  2. L23:2010 excites
  3. L19:1036 suppresses
  4. L22:978 suppresses
  5. L21:2772 excites
  6. L22:144 suppresses
  7. L20:54 suppresses
  8. L17:118 excites comma after a clause ending
  9. L18:2795 suppresses
  10. L24:35 suppresses
  11. L20:541 weakly suppresses
  12. L14:35 weakly suppresses
  13. L19:400 weakly suppresses
  14. L18:119 weakly excites
  15. L18:178 weakly suppresses
  16. L24:985 weakly suppresses
  17. L15:65 weakly suppresses
  18. L17:473 weakly excites
  19. L21:583 weakly suppresses
  20. L10:39 weakly excites
  21. L6:16 weakly excites
  22. L16:1112 weakly suppresses
  23. L19:547 weakly excites
  24. L8:2539 weakly suppresses

Feeds later layers

  1. L27:719 strongly excites
  2. L27:878 strongly excites
  3. L26:820 strongly excites
  4. L27:996 excites
  5. L27:1000 suppresses
  6. L26:1854 excites
  7. L27:2982 weakly excites
  8. L26:1041 weakly excites
  9. L27:509 weakly excites
  10. L26:416 weakly excites
  11. L26:1989 weakly suppresses
  12. L26:1147 weakly excites
  13. L27:1403 weakly excites
  14. L26:1577 weakly excites
  15. L27:2425 weakly suppresses
  16. L26:2011 weakly suppresses
  17. L26:1728 weakly excites
  18. L27:1582 weakly excites
  19. L26:2626 weakly excites
  20. L27:1987 weakly excites
  21. L26:948 weakly suppresses
  22. L26:581 weakly excites
  23. L27:2871 weakly excites
  24. L27:2927 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.

3 values between -7.65 and -6.572 values between -6.57 and -5.486 values between -5.48 and -4.4020 values between -4.40 and -3.3215 values between -3.32 and -2.2471 values between -2.24 and -1.15831 values between -1.15 and -0.072626 values between -0.07 and 1.01144 values between 1.01 and 2.1050 values between 2.10 and 3.1830 values between 3.18 and 4.2625 values between 4.26 and 5.3425 values between 5.34 and 6.4312 values between 6.43 and 7.514 values between 7.51 and 8.5911 values between 8.59 and 9.6711 values between 9.67 and 10.764 values between 10.76 and 11.841 values between 11.84 and 12.922 values between 12.92 and 14.002 values between 14.00 and 15.092 values between 17.25 and 18.332 values between 19.42 and 20.502 values between 20.50 and 21.582 values between 21.58 and 22.671 values between 22.67 and 23.752 values between 23.75 and 24.831 values between 29.16 and 30.241 values between 34.57 and 35.66offnegativepositiveoften

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. i
  2. i
  3. *i
  4. →i
  5. {i
  6. ,i
  7. )i
  8. =i
  9. (i
  10. .i

Suppresses

  1. riba
  2. avras
  3. unter
  4. atrás
  5. ético
  6. uebas
  7. ayers
  8. aine
  9. auses
  10. éc