Neuron Explorer

Qwen3-0.6B

'for' introducing a purpose or target

A neuron in layer 16 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 the word 'for' when it points to a purpose or recipient, like 'eBook for Free', 'activities for the book', or 'for seven days'. Its negative side fires on 'there' in 'as many ... as there are'. Steering output was unrelated text, so the pattern rests on the snippets alone.

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

    In this report, the authors conclude with implications for policymakers and LEAs.⏎Challenges in Implementing Physical Security Measures in K–12 Schools⏎Download eBook for Free⏎|PDF file||0.6 MB||⏎Use Adobe Acrobat Reader version

  2. #2 peak

    a perfect enhancement to E.L. Konigsburg's Newberry Medal Winner.⏎This unit contains vocabulary, comprehension and writing activities for the book, The Magic School Bus: a Hurricane by Joanna Cole. Common Core

  3. #3 peak

    , if errno is non-zero, an error has occurred. The ldexpf() and ldexpl() functions do not set errno.⏎See attributes(5) for descriptions of the following attributes:<|endoftext|>The

  4. #4 peak

    you will be asked to critically think and debate varying interpretations of the empirical evidence in relation to the science being brought to your attention<|endoftext|>Economics– Important Questions Bank for Rajasthan Board HSC 201

  5. #5 peak

    recommend or endorse any third party, product, service or information provided on this site.⏎If you have dietary restrictions and/or allergies, always read the ingredient list carefully for all products prior to consumption. If the ingredient list is not available on the product

  6. #6 peak

    . Nicklin.”⏎Israel Thomas Nicklin was born in Loudon County, Virginia, August 12, 1797. After attending Fairfax School for several years, he moved with his parents to Bowles Mills, Tyler County,

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: the first's, first's, and first's. I think it's going to be a first's. Maybe's. Maybe's. Maybe's. So, maybe's. But, maybe's. Maybe's. Wait, maybe's

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

    the pinnacle and its prize, only to lose their footing and slide back down to their harm or death.⏎How many glass mountains are there? Maybe as many as there are people. For me, the most devious of glass mountains appears in the

  2. #2 peak

    are 221293000 people.⏎Pakistan telephone code is 92.<|endoftext|>T HERE are as many definitions of public opinion as there aredisciplinary points of view. Political scientists see public opinion interms of the distribution

  3. #3 peak

    ⏎Review: Brave. A Book About Courage⏎There are as many definitions of poetry as there are poets. Poetry is an artistic expression unlike any other. Some research suggests that

  4. #4 peak

    residence itself, or a brook falling down the sward between trees and bushes and clumpy growths. There are as many forms and kinds of gardens as there are persons who have gardens […].” –LHBGC, p.

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. L5:858 strongly excites
  2. L1:47 excites
  3. L2:1866 suppresses
  4. L0:41 suppresses
  5. L1:32 excites
  6. L0:2724 suppresses
  7. L3:381 excites 'for' in research institute names
  8. L10:1473 weakly suppresses
  9. L15:1587 weakly excites
  10. L2:1925 weakly suppresses
  11. L15:792 weakly suppresses
  12. L0:1543 weakly excites
  13. L0:564 weakly excites
  14. L0:2765 weakly excites
  15. L3:1055 weakly suppresses
  16. L4:189 weakly excites
  17. L15:1135 weakly suppresses
  18. L11:2595 weakly suppresses
  19. L15:1616 weakly excites word fragments mid-word
  20. L1:2601 weakly suppresses
  21. L15:201 weakly excites
  22. L12:2810 weakly suppresses
  23. L15:1900 weakly suppresses
  24. L0:2211 weakly suppresses

Feeds later layers

  1. L18:650 suppresses the word 'more' before a comparison
  2. L27:2224 suppresses
  3. L17:253 suppresses
  4. L19:572 suppresses
  5. L20:601 suppresses
  6. L21:812 suppresses
  7. L19:1070 suppresses
  8. L19:1237 suppresses
  9. L19:1382 excites 'addition' in 'In addition to'
  10. L23:1368 suppresses
  11. L20:934 excites
  12. L26:1286 suppresses
  13. L21:2717 excites
  14. L22:2883 suppresses
  15. L18:778 excites
  16. L17:1928 excites
  17. L19:1728 weakly suppresses
  18. L19:2221 weakly excites
  19. L26:2860 weakly suppresses
  20. L22:1141 weakly excites
  21. L20:2375 weakly excites
  22. L26:2583 weakly suppresses
  23. L18:687 weakly excites
  24. L25:2511 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 -5.50 and -5.311 values between -5.31 and -5.121 values between -5.12 and -4.931 values between -4.93 and -4.741 values between -4.74 and -4.541 values between -4.54 and -4.351 values between -4.35 and -4.162 values between -3.97 and -3.781 values between -3.78 and -3.592 values between -3.59 and -3.401 values between -3.40 and -3.212 values between -3.21 and -3.022 values between -3.02 and -2.834 values between -2.83 and -2.645 values between -2.64 and -2.454 values between -2.45 and -2.268 values between -2.26 and -2.0710 values between -2.07 and -1.8810 values between -1.88 and -1.6922 values between -1.69 and -1.5024 values between -1.50 and -1.3148 values between -1.31 and -1.1253 values between -1.12 and -0.9371 values between -0.93 and -0.74126 values between -0.74 and -0.55201 values between -0.55 and -0.36346 values between -0.36 and -0.17925 values between -0.17 and 0.031036 values between 0.03 and 0.22643 values between 0.22 and 0.41275 values between 0.41 and 0.6059 values between 0.60 and 0.7910 values between 0.79 and 0.982 values between 0.98 and 1.174 values between 1.17 and 1.362 values between 1.36 and 1.552 values between 1.74 and 1.931 values between 1.93 and 2.12offnegativepositiveoften

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. sg
  2. e
  3. 冲动
  4. \xd
  5. \xb
  6. olatile
  7. ingular
  8. escription
  9. arrow
  10. ⎝

Suppresses

  1. iously
  2. ium
  3. bij
  4. ialis
  5. liable
  6. 镡
  7. anium
  8. thur
  9. rett
  10. izont