Neuron Explorer

Qwen3-0.6B

'to' and 'of' inside phrases

A neuron in layer 0 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron responds to small connecting words like 'to' and 'of' that sit in the middle of a phrase, often after a verb or noun. It goes quiet, or slightly negative, on fragments of longer words such as the end of 'alleys'. Turning it up or down did not change the writing in a way that matches this pattern.

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

    ,<|endoftext|>This material must not be used for commercial purposes, or in any hospital or medical facility. Failure to comply may result in legal action.⏎WHAT YOU NEED TO KNOW:⏎Diphtheria is a disease caused by a bacterial infection. The

  2. #2 peak

    technique in the opening scene, when Enfield relates his story of witnessing Hyde trample a little girl—a night when the streets were so empty that he began “to long for the sight of a policeman.”⏎This notion of the city as a fearful

  3. #3 peak

    HAVE DIFFERENT REACTION RATES BECAUSE OF DIFFERENT STRUCTURE OF THE COMPOUNDS. DIFFERENT REACTION WILL YIELDS TO A CONCLUSION OF WHAT IS/ARE THE FACTOR/S OF ELECTROPH

  4. #4 peak

    two component molecules, and will⏎reassociate upon slow cooling under appropriate salt conditions. So,⏎the issue is not decomposition so much as disassociation upon exposure⏎to too high a temperature; and the reason for the disassociation is that⏎the

  5. #5 peak

    genes has led to the identification of highly conserved consensus sequences⏎at the 5' and 3' ends of essentially all mRNA introns.(Figure⏎of Consensus Splice Sites) A majority of these searches are based on algorithms that

  6. #6 peak

    ⏎the answer may not have known in fact if someone else had been naturalized.⏎An individual may have said yes because he felt it was the right thing⏎to say or he intended to begin the process.⏎-

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: la question de liserie, le liserie, le liserie, la liserie, la liserie, la liserie, la liserie, la liserie, la liserie, la

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

    promoters, Ace Borger and J.T. Peyton, were engaged. The land was then surveyed, subdivided into blocks, lots, streets, avenues, and alleys. The Stinnett Town site Company dedicated the streets, avenues and alle

  2. #2 peak

    Rights Division⏎A Guide To Federal Religious Land Use Protections⏎The Religious Land Use and Institutionalized Persons Act (RLUIPA)⏎protects religious institutions from und

  3. #3 peak

    of errors in the street grid, labeling Locust Street as “Lock Street” and Grand Avenue as “Main Street.” Nor does he depict the sixteen-foot-wide alleys that are clearly platted on the Sanborn maps, although in several cases

  4. #4 peak

    linear, curving hills and depressions, which are most distinct in the right of the image. Many narrow, linear grooves run in various directions across this undulating terrain. There are some small, less than 1 kilometer (0

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

None: this is the first layer.

Feeds later layers

  1. L1:341 suppresses
  2. L3:7 suppresses
  3. L2:21 excites
  4. L1:43 weakly excites
  5. L1:571 weakly suppresses
  6. L16:253 weakly suppresses
  7. L1:150 weakly excites
  8. L2:17 weakly excites
  9. L17:385 weakly excites
  10. L17:852 weakly excites
  11. L16:890 weakly suppresses
  12. L2:407 weakly excites
  13. L1:862 weakly excites
  14. L17:2014 weakly excites
  15. L22:2455 weakly excites
  16. L23:2691 weakly excites
  17. L3:504 weakly excites
  18. L23:312 weakly suppresses
  19. L26:2676 weakly suppresses
  20. L1:232 weakly excites
  21. L10:301 weakly excites
  22. L17:707 weakly suppresses
  23. L19:1020 weakly suppresses
  24. L3:722 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.

2 values between -1.00 and -0.961 values between -0.87 and -0.831 values between -0.74 and -0.691 values between -0.69 and -0.651 values between -0.56 and -0.521 values between -0.52 and -0.471 values between -0.47 and -0.431 values between -0.43 and -0.385 values between -0.38 and -0.341 values between -0.34 and -0.291 values between -0.29 and -0.2510 values between -0.25 and -0.2024 values between -0.20 and -0.1657 values between -0.16 and -0.12133 values between -0.12 and -0.07402 values between -0.07 and -0.032365 values between -0.03 and 0.02612 values between 0.02 and 0.06159 values between 0.06 and 0.1162 values between 0.11 and 0.1539 values between 0.15 and 0.2012 values between 0.20 and 0.247 values between 0.24 and 0.282 values between 0.28 and 0.333 values between 0.37 and 0.421 values between 0.42 and 0.461 values between 0.46 and 0.511 values between 0.60 and 0.641 values between 0.64 and 0.681 values between 0.73 and 0.77offnegativepositiveoften

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. 菟
  2. 恚
  3. well
  4. ięć
  5. perfection
  6. _cookies
  7. TERS
  8. tep
  9. numbering
  10. 完美

Suppresses

  1. ingly
  2. 主人
  3. <>⏎
  4. /task
  5. /query
  6. quina
  7. /event
  8. halb
  9. ös
  10. ogenous