Neuron Explorer

Qwen3-0.6B

biblical quotations and scripture

A neuron in layer 11 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron tends to light up inside quoted passages from the Bible and other religious or moralizing text, such as words of Jesus or Proverbs. The steering tests did not show a clear theme, so this is only a rough guess.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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 "cheating minister", his sin gives him "sympathies so intimate with the sinful brotherhood of mankind, so that his chest vibrate[s] in unison with theirs."⏎His eloquent and powerful sermons derive from

  2. #2 peak

    heard was a law. God once told Jacob (Gen. 26:4-5), “I will make your seed to multiply as the stars of heaven, and I will give to your seed all these countries. And in your seed

  3. #3 peak

    the way which I command you, that it may be well with you.'⏎Matthew 7:24-27 “Therefore everyone who hears these words of Mine and acts on them, may

  4. #4 peak

    Holman Christian Standard BibleJesus told him, “Foxes have dens, and birds of the sky have nests, but the Son of Man has no place to lay His head.”⏎New American Standard BibleAnd Jesus

  5. #5 peak

    14). As one historian characterized this trans-Atlantic movement, it attempted “to limit the socially destructive effects of morally unhindered capitalism, to extract from those [capitalist] markets the tasks they had demonstrably bungled, to counter

  6. #6 peak

    questioning salt guidelines riles heart experts⏎- CDC: 2012 was deadliest year for West Nile in US⏎- What stresses moms most? Thems

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 if I, like a new person, have been able to create a more transparent and authentic version of myself, like a persona, which I call "I" (or whatever) I? This is a new identity, more transparent,

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

    Geriatric Medicine study are available here: http://www.europeangeriaticmedicine.com/article/S1878-7649(13)00123-X/).⏎It is possible nursing staff underestimate the importance

  2. #2 peak

    2 km. The infrasound data is collected at 20 hz, while the meteorologic data is recorded at 1 hz.⏎Figure 4. Array layout. Blue dots indicate the 7 elements, while the purple

  3. #3 peak

    intent as the underlying mechanism. Journal of Marketing, 18(1), 39-53. http://dx.doi.org/10.1080/13527266.20

  4. #4 peak

    . Journal of Business Research, 67(8), 1595-1600. http://dx.doi.org/10.1016/j.jbusres.2013.10

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. L10:2958 suppresses
  2. L9:1461 suppresses
  3. L2:112 weakly excites
  4. L8:232 weakly suppresses
  5. L8:1332 weakly suppresses
  6. L7:10 weakly suppresses
  7. L5:2215 weakly excites
  8. L0:1100 weakly excites
  9. L1:1740 weakly suppresses
  10. L7:865 weakly suppresses
  11. L8:1088 weakly suppresses
  12. L8:2036 weakly suppresses
  13. L5:512 weakly excites
  14. L6:1661 weakly suppresses
  15. L5:930 weakly excites
  16. L8:1785 weakly suppresses
  17. L9:2857 weakly excites
  18. L6:1760 weakly excites data quality and data formatting
  19. L2:52 weakly excites
  20. L3:1796 weakly suppresses
  21. L9:2518 weakly excites
  22. L3:1053 weakly excites
  23. L10:446 weakly excites
  24. L6:2899 weakly suppresses

Feeds later layers

  1. L16:2040 weakly excites
  2. L15:2289 weakly excites
  3. L26:719 weakly excites
  4. L26:2777 weakly suppresses
  5. L21:2712 weakly excites
  6. L19:1663 weakly excites
  7. L21:188 weakly excites
  8. L26:476 weakly suppresses
  9. L19:2678 weakly suppresses
  10. L18:2199 weakly suppresses
  11. L25:2264 weakly suppresses
  12. L13:1540 weakly excites
  13. L15:929 weakly excites
  14. L27:226 weakly suppresses
  15. L20:2246 weakly excites
  16. L21:2403 weakly excites
  17. L15:1515 weakly excites
  18. L18:2070 weakly excites
  19. L18:2307 weakly suppresses
  20. L24:1793 weakly suppresses
  21. L21:872 weakly excites
  22. L24:2248 weakly suppresses
  23. L17:2472 weakly suppresses
  24. L27:277 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 -3.39 and -3.271 values between -1.34 and -1.232 values between -1.00 and -0.891 values between -0.89 and -0.773 values between -0.77 and -0.665 values between -0.66 and -0.5411 values between -0.54 and -0.4335 values between -0.43 and -0.32127 values between -0.32 and -0.20501 values between -0.20 and -0.092010 values between -0.09 and 0.02875 values between 0.02 and 0.14243 values between 0.14 and 0.2562 values between 0.25 and 0.3713 values between 0.37 and 0.489 values between 0.48 and 0.596 values between 0.59 and 0.711 values between 0.82 and 0.931 values between 0.93 and 1.051 values between 1.05 and 1.16offnegativepositiveoften

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. nü
  2. theless
  3. 但不限
  4. 在地上
  5. /Area
  6. ":-
  7. istribute
  8. ":"'
  9. uninstall
  10. ẩu

Suppresses

  1. -
  2. (
  3. volt
  4. *v
  5. **
  6. 值
  7. *
  8. *j
  9. |
  10. value