Neuron Explorer

Qwen3-0.6B

words after 'as though' or 'as if'

A neuron in layer 7 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 word right after comparison phrases like 'as though' or 'as if', and on a few related words like 'also' and 'that' in advice sentences. It is a loose pattern. Turning it down or up changed the model's topic without a clear link to these phrases.

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

    book:⏎The two entities that were supposed to write the regulations, the Pay Board and the Price Commission, wrangled and dithered. When it looked as though they were going to miss a crucial deadline

  2. #2 peak

    obvious effect on prospective memory, the study’s authors recommend changing the prevailing view that the allele has only subtle, often undetectable effects on cognition. The findings also supplement previous discoveries of how the allele also is linked to problems in epis

  3. #3 peak

    one per student)⏎-four large pieces of paper or poster board, each with one of the following statements written in large letters: “Strongly Agree,” “Agree Somewhat,” “Disagree Somewhat,” “Strongly Disagree”⏎Activities

  4. #4 peak

    cool at 5 degrees C or below can prevent germs from growing and multiplying.⏎2. Ensure Food is stored at the right temperature⏎It is also recommended that you do not store warm or hot food items directly into the fridge. The sudden

  5. #5 peak

    without debugging (Ctrl+F5)⏎- For any runtime error, you can debug the program as follows:⏎- Select Debug >> Start debugging (F5)⏎- Please note that during debugging, the program will show and close the console display immediately.⏎

  6. #6 peak

    left is both a __________________ and __________________.⏎Your brain also solves mysteries in language. Imagine you find the cryptic note to the right on a food container. Good thing your brain needs only the first and last letter of a word to be in the

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 meaning of the line in the middle of a circle in the context of the problem, and how does it relate to the problem of the time of the day?" I need to think about the problem of the time of the

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

    as a castaway among the native inhabitants of pre-colonial Cuba.⏎María Estrada (the surname is given as Destrada or Estrada in some sources) was born in Seville, although her father came originally from northern Spain

  2. #2 peak

    that are called whenever⏎a child is launched or terminated. (The Win32 version of Apache uses multithreading⏎rather than a multiprocess model, but as of this writing modules are not given⏎a chance to take action when a

  3. #3 peak

    to make an image of a 1GB drive, and store it on another hard drive in the same PC.⏎Note: always be cautious when using dd, as it’s very easy to completely wipe out a drive, as we will show later

  4. #4 peak

    irth: Or Pills to Purge Melancholy’ by Thomas D’Urfey in 1719 (though ‘Wit and Mirth’ was originally published in<|endoftext|>This week's mental

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. L6:1086 weakly excites
  2. L2:472 weakly suppresses
  3. L5:302 weakly suppresses
  4. L0:2814 weakly excites
  5. L5:1710 weakly excites
  6. L6:2211 weakly suppresses
  7. L6:2040 weakly suppresses
  8. L6:966 weakly excites
  9. L5:1349 weakly suppresses
  10. L1:1018 weakly suppresses
  11. L5:1914 weakly excites
  12. L6:2439 weakly excites
  13. L6:1883 weakly suppresses
  14. L4:632 weakly excites
  15. L2:335 weakly excites
  16. L4:63 weakly excites
  17. L1:777 weakly excites
  18. L5:887 weakly suppresses
  19. L4:1736 weakly suppresses
  20. L3:2046 weakly excites
  21. L1:1528 weakly suppresses
  22. L5:1451 weakly suppresses
  23. L2:3021 weakly suppresses
  24. L4:570 weakly suppresses

Feeds later layers

  1. L8:362 suppresses unclear: evaluative or explanatory phrase words
  2. L8:739 excites
  3. L8:1723 suppresses
  4. L8:480 suppresses
  5. L13:2748 excites
  6. L10:2511 suppresses
  7. L9:118 excites
  8. L8:1950 suppresses
  9. L12:118 suppresses
  10. L10:1487 excites
  11. L8:2329 excites
  12. L11:1628 weakly suppresses
  13. L8:1291 weakly excites
  14. L9:1943 weakly suppresses
  15. L8:597 weakly excites
  16. L12:376 weakly suppresses
  17. L8:44 weakly excites
  18. L8:510 weakly excites
  19. L10:1123 weakly excites
  20. L12:861 weakly suppresses
  21. L13:385 weakly excites
  22. L10:1310 weakly suppresses
  23. L8:50 weakly excites
  24. L8:91 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 -1.64 and -1.562 values between -1.56 and -1.471 values between -1.47 and -1.391 values between -1.39 and -1.312 values between -1.31 and -1.235 values between -1.23 and -1.143 values between -1.14 and -1.063 values between -1.06 and -0.983 values between -0.98 and -0.904 values between -0.90 and -0.814 values between -0.81 and -0.737 values between -0.73 and -0.6513 values between -0.65 and -0.5712 values between -0.57 and -0.4818 values between -0.48 and -0.4036 values between -0.40 and -0.3274 values between -0.32 and -0.24194 values between -0.24 and -0.15449 values between -0.15 and -0.071542 values between -0.07 and 0.01829 values between 0.01 and 0.09294 values between 0.09 and 0.18150 values between 0.18 and 0.2680 values between 0.26 and 0.3466 values between 0.34 and 0.4227 values between 0.42 and 0.5120 values between 0.51 and 0.5917 values between 0.59 and 0.6714 values between 0.67 and 0.7511 values between 0.75 and 0.846 values between 0.84 and 0.928 values between 0.92 and 1.005 values between 1.08 and 1.171 values between 1.17 and 1.253 values between 1.25 and 1.331 values between 1.33 and 1.421 values between 1.50 and 1.581 values between 1.58 and 1.66offnegativepositiveoften

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. ozo
  2. 那你
  3. aser
  4. 蹼
  5. roupe
  6. pomocą
  7. Rosenstein
  8. ’est
  9. rase
  10. 报记者

Suppresses

  1. *
  2. *t
  3. "*
  4. че
  5. σ
  6. icum
  7. schedules
  8. *s
  9. overhead
  10. *n