Neuron Explorer

Qwen3-0.6B

comparison targets like 'than Earth'

A neuron in layer 14 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 fires on the thing being compared to, such as 'Earth' in 'larger than Earth' or 'much like Earth'. Turning it up gave text that did not obviously relate to comparisons.

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

    with a venting parent. As a result, they might have learned to suppress their anger for fear of retaliation or made a personal vow never to be like their parent.⏎The venter style is at the opposite end of the anger management continuum.

  2. #2 peak

    exoplanets — could also be rocky and have a composition much like Earth’s. Some might even harbor life. Over 99% of those alien

  3. #3 peak

    coil that induces them to combine into helium atoms. The process also releases neutrons.⏎The Fusion Engine provides energy in two ways. Like in a fission reactor, the energy of the scattered neutrons gives off heat that ultimately drives a turbine.

  4. #4 peak

    , built by Airbus, of methane and carbon dioxide in the atmosphere of K2-18b, an exoplanet 8.6 times bigger than Earth, shows there is still a lot to uncover in the search for habitable environments

  5. #5 peak

    determined autocrat but not an antisemite. Portugal’s very small and then only recently reestablished Jewish community possessed the same rights as the country’s other citizens. It retained

  6. #6 peak

    to discover planets ranging from huge gas giants, some of them orbiting close to their stars, to “super Earths,” Earth-like planets that are larger than Earth but still might contain life.<|endoftext|>New guidelines have been launched for educators, using

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: something that has to do with something that has to do with something else... something... something... something... what... what... what... what... what... what... what... what... what... what... what... what... what... what

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

    striking; certain passages of the hadeeth could easily be mistaken for discussions about contemporary environmental issues.⏎The Prophet sall Allaahu`alayhi wa sallam's environmental philosophy is first of all holistic: it assumes a fundamental link and inter

  2. #2 peak

    GNU package:octave⏎released on 31 May 2012⏎|License||Verified by||Verified on||Notes|⏎|GPLv2||Kelly Hopkins||6 December 2011|⏎|GPLv3

  3. #3 peak

    ious balance between man and nature. From all accounts of his life and deeds, we read that the Prophet sall Allaahu`alayhi wa sallam had a profound respect for fauna and flora, as well as an almost visceral connection

  4. #4 peak

    to teach you to observe in the land that you are crossing the Jordan to possess, so that you, your children and their children after them may fear the Lord your God as long as you live by keeping all his decrees and commands that I

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. L6:982 excites
  2. L3:577 excites
  3. L10:2125 suppresses
  4. L2:291 weakly suppresses
  5. L12:2772 weakly suppresses comma after an introductory phrase
  6. L2:937 weakly suppresses
  7. L7:2418 weakly suppresses
  8. L11:708 weakly suppresses
  9. L12:625 weakly excites
  10. L8:1259 weakly suppresses
  11. L11:525 weakly suppresses parentheses and code fragments
  12. L8:971 weakly suppresses
  13. L3:2884 weakly suppresses
  14. L12:1021 weakly suppresses
  15. L11:2214 weakly excites
  16. L11:2568 weakly excites
  17. L12:1040 weakly excites
  18. L4:2499 weakly suppresses
  19. L8:868 weakly excites
  20. L13:528 weakly excites
  21. L9:1945 weakly suppresses
  22. L5:2896 weakly suppresses
  23. L8:316 weakly excites
  24. L1:157 weakly suppresses

Feeds later layers

  1. L17:2737 weakly excites
  2. L22:702 weakly excites
  3. L18:1822 weakly excites
  4. L19:2955 weakly excites
  5. L15:452 weakly excites
  6. L24:439 weakly excites
  7. L21:2726 weakly excites
  8. L17:1942 weakly suppresses
  9. L24:532 weakly excites
  10. L18:2303 weakly excites
  11. L15:401 weakly excites
  12. L21:1041 weakly suppresses
  13. L15:1470 weakly excites
  14. L18:2441 weakly suppresses
  15. L21:2046 weakly excites
  16. L20:246 weakly excites
  17. L25:1172 weakly excites
  18. L18:2631 weakly excites
  19. L23:1334 weakly excites
  20. L16:789 weakly suppresses
  21. L21:1405 weakly excites
  22. L18:2117 weakly excites
  23. L15:2956 weakly excites
  24. L18:2459 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 -2.27 and -2.112 values between -1.33 and -1.181 values between -1.18 and -1.023 values between -1.02 and -0.8610 values between -0.86 and -0.7117 values between -0.71 and -0.5524 values between -0.55 and -0.4093 values between -0.40 and -0.24411 values between -0.24 and -0.082545 values between -0.08 and 0.07600 values between 0.07 and 0.23116 values between 0.23 and 0.3935 values between 0.39 and 0.5420 values between 0.54 and 0.708 values between 0.70 and 0.854 values between 0.85 and 1.015 values between 1.01 and 1.171 values between 1.17 and 1.324 values between 1.32 and 1.481 values between 1.48 and 1.641 values between 1.64 and 1.792 values between 1.79 and 1.951 values between 1.95 and 2.102 values between 2.10 and 2.261 values between 3.82 and 3.98offnegativepositiveoften

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. explicit
  2. 隅
  3. formance
  4. 舞台上
  5. 奔跑
  6. (Key
  7. 适时
  8. 早点
  9. Am
  10. ListAdapter

Suppresses

  1. grown
  2. 系
  3. essel
  4. 故
  5. itus
  6. inent
  7. "-//
  8. i
  9. 洗净
  10. uelle