Neuron Explorer

Qwen3-0.6B

opening parenthesis in formulas and citations

A neuron in layer 10 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 starts a formula term or a copyright notice, such as 'Copyright (c)' or 'n = N / (1 + ...)'. The activations are fairly weak. Turning it up did not produce a clear theme in the generated text.

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

    the behavior of a population is known at all. * How to Use Slovin's Formula⏎Slovin's formula is written as:⏎n = N / (1 + Ne^2)⏎n = Number of samples⏎N = Total population

  2. #2 peak

    of the two overloaded versions provided in C++.⏎/* memchr example */⏎int main ()⏎char * pch;⏎char str = "Example string";⏎pch = (char

  3. #3 peak

    water, were the major factors shaping phytoplankton communities.⏎- Abstract views: 1777⏎- PDF: 842⏎Copyright (c) 2009 Silvia TAVERNINI, Daniele N

  4. #4 peak

    viable. A symbol or instrument found in the Tuturangiana Andala ritual of green bamboo, betel nut, black goat, and red flag.⏎Copyright (c) 2022 Proceedings Of International Conference On Communication Science⏎This

  5. #5 peak

    Here are examples of creating markers, setting markers, and moving point to markers:⏎;; Make a new marker that initially does not point anywhere: (setq m1 (make-marker)) ⇒ #<marker in no buffer>⏎m1to point between

  6. #6 peak

    leaves to refer to their objects⏎> through a Map. Then changing a leaf object requires only that the value⏎> associated with key represented by the leaf be (re-)inserted into the⏎> Map. The Tree itself need not change

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 it, and I feel it. I have been thinking about it, and I feel it. I feel, I- I- I- I- I- I- I- I- ... I- I-

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

    Satellite data can help scientists understand:⏎- The CO2 pump: remote sensing of phytoplankton (green plant-life) in the ocean.⏎- Sea Surface Temperature variation: which shows the temperature variation, dynamics and some chaos – as shown

  2. #2 peak

    behavior of a population is known at all. * How to Use Slovin's Formula⏎Slovin's formula is written as:⏎n = N / (1 + Ne^2)⏎n = Number of samples⏎N = Total population⏎

  3. #3 peak

    ave and another convex, which it termed perspicillum: was born the first astronomical telescope refractor. The manufacturer of lenses designed it with the idea of observing distant objects on the Earth's surface,

  4. #4 peak

    risk factors. Mark Goldstein, 79,<|endoftext|>Sea surface temperature⏎Sea surface temperature (SST) is the water temperature at the surface. In practical terms

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. L8:3 excites
  2. L6:144 excites
  3. L9:796 suppresses
  4. L8:7 excites
  5. L8:12 excites
  6. L3:464 weakly excites
  7. L8:13 weakly suppresses
  8. L8:1765 weakly excites
  9. L2:28 weakly excites
  10. L8:2143 weakly excites
  11. L8:16 weakly suppresses
  12. L3:843 weakly suppresses
  13. L5:1993 weakly excites
  14. L9:1532 weakly suppresses
  15. L7:288 weakly excites
  16. L3:1693 weakly excites
  17. L6:43 weakly suppresses
  18. L7:2665 weakly suppresses
  19. L8:10 weakly excites
  20. L5:2107 weakly suppresses
  21. L7:76 weakly excites
  22. L4:2793 weakly suppresses
  23. L8:2776 weakly excites
  24. L4:779 weakly suppresses

Feeds later layers

  1. L16:995 strongly excites
  2. L14:1318 excites "both" and "either" before alternatives
  3. L11:1272 excites unclear: sentence ends in factual prose
  4. L14:1274 excites
  5. L11:2310 excites opening parenthesis in formulas and math
  6. L19:1497 weakly excites
  7. L15:2042 weakly excites
  8. L16:2034 weakly suppresses
  9. L13:2444 weakly excites
  10. L15:801 weakly excites 'of' inside organisation names
  11. L13:2590 weakly excites
  12. L13:1022 weakly excites
  13. L18:217 weakly excites
  14. L15:1794 weakly excites
  15. L13:2903 weakly excites
  16. L18:1459 weakly suppresses
  17. L23:2411 weakly suppresses
  18. L26:1018 weakly excites
  19. L16:1513 weakly excites opening parenthesis before a numbered list
  20. L19:2390 weakly suppresses
  21. L22:246 weakly suppresses
  22. L17:3071 weakly suppresses
  23. L13:1698 weakly excites
  24. L15:24 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.

1 values between -0.72 and -0.691 values between -0.69 and -0.663 values between -0.54 and -0.514 values between -0.51 and -0.484 values between -0.48 and -0.441 values between -0.44 and -0.413 values between -0.41 and -0.384 values between -0.38 and -0.354 values between -0.35 and -0.3212 values between -0.32 and -0.2915 values between -0.29 and -0.2623 values between -0.26 and -0.2346 values between -0.23 and -0.2059 values between -0.20 and -0.1788 values between -0.17 and -0.14143 values between -0.14 and -0.11223 values between -0.11 and -0.08373 values between -0.08 and -0.05704 values between -0.05 and -0.021004 values between -0.02 and 0.02555 values between 0.02 and 0.05260 values between 0.05 and 0.08140 values between 0.08 and 0.1187 values between 0.11 and 0.1453 values between 0.14 and 0.1731 values between 0.17 and 0.2021 values between 0.20 and 0.2317 values between 0.23 and 0.2610 values between 0.26 and 0.295 values between 0.29 and 0.328 values between 0.32 and 0.351 values between 0.35 and 0.381 values between 0.38 and 0.411 values between 0.41 and 0.442 values between 0.44 and 0.471 values between 0.47 and 0.50offnegativepositiveoften

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. rias
  2. otos
  3. \",\
  4. obar
  5. pto
  6. airo
  7. 专线
  8. rios
  9. éri
  10. etAddress

Suppresses

  1. onnen
  2. idelity
  3. わけではない
  4. ╝
  5. almart
  6. borg
  7. ises
  8. ству
  9. .instance
  10. esting