Neuron Explorer

Qwen3-0.6B

'a' after a preposition

A neuron in layer 9 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 'a' or 'an' when it comes after a preposition like 'for', 'from' or 'of', as in 'look out for a definition'. Its opposite is 'the' after 'one of'. Turning it up did not produce a clear theme in the generated text.

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 general, but if they get in this habit early, they will be building a routine that can last a lifetime. Ask what your child wants to do for a

  2. #2 peak

    the panelists could come to a common understanding. This raised a few questions in my head. Disappointed with the expert advice, I decided to look out for a definition of my own.⏎As what’s needed in such a situation, I looked

  3. #3 peak

    ill-equipped men in his command. Fremont, whose Federals were also ill-equipped and trained, had nearly 40,000.⏎Price left a small garrison at Lexington, which was overrun by Major Frank J. White’s

  4. #4 peak

    .⏎We tend to dwell on the deaths, but often overlook the survivors of a life altering collision. Crashes don’t discriminate! Everyone is at risk, from a newborn properly restrained in a car seat, to a bicyclist or pedestrian in the

  5. #5 peak

    Hitler on both occasions, thus putting himself firmly on the Gestapo’s list of enemies; in 1938, as the Nazi vice was closing on an independent Austria, the archduke, at obvious risk to his life, volunteered

  6. #6 peak

    System that contains information on reported live births in the U.S.⏎The researchers were interested in pinning down the average age of first-time mothers since the age of a woman’s first birth has a bearing on the total number of births she may have

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: "I don't know about a date, but I have to go on the record." So, I'm not sure about anything, but I have to go on the record. [Inception] 04.12.

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

    number of cancers and scientists around the globe are busy trying to harness the power of the immune system in the fight against cancer. The HPV vaccine is just one of the first; expect to hear more and more about the topic in the future.⏎The

  2. #2 peak

    . As you develop your<|endoftext|>Catherine Rampell had a post last week declaring that “College is (Still) Worth It“. The piece is the latest in a series that she and her Economix blog colleague

  3. #3 peak

    , the United States has lost over 50 percent of its managed honeybee colonies over the past 10 years.⏎Pollinator gardens are just one of the many ways to provide both shelter and food to insects teetering on the brink

  4. #4 peak

    This video is part two of a seven-part National Academies series, Climate Change: Lines of Evidence. The video

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. L7:312 excites
  2. L6:61 suppresses
  3. L8:636 suppresses
  4. L6:29 suppresses
  5. L5:125 excites "of" after all three, both, ten
  6. L3:528 excites
  7. L8:475 excites
  8. L7:396 excites
  9. L3:47 weakly excites
  10. L4:217 weakly suppresses
  11. L4:16 weakly suppresses
  12. L3:489 weakly suppresses
  13. L8:1232 weakly excites
  14. L1:235 weakly suppresses
  15. L6:819 weakly suppresses
  16. L8:508 weakly excites
  17. L1:2074 weakly suppresses
  18. L3:545 weakly suppresses
  19. L7:2164 weakly suppresses period inside country abbreviations
  20. L4:731 weakly excites
  21. L8:333 weakly excites
  22. L7:401 weakly suppresses
  23. L4:48 weakly excites
  24. L3:472 weakly suppresses

Feeds later layers

  1. L11:1280 excites
  2. L14:655 excites
  3. L14:169 excites
  4. L18:381 excites
  5. L15:716 excites
  6. L11:55 suppresses
  7. L18:172 suppresses
  8. L14:600 excites
  9. L12:652 excites
  10. L14:871 excites
  11. L15:1205 excites
  12. L17:362 excites
  13. L18:715 weakly suppresses
  14. L13:976 weakly suppresses
  15. L12:2478 weakly excites
  16. L17:146 weakly suppresses period after abbreviations and dates
  17. L15:37 weakly excites
  18. L20:251 weakly suppresses
  19. L16:66 weakly suppresses
  20. L22:194 weakly suppresses
  21. L16:528 weakly excites
  22. L14:1876 weakly excites
  23. L21:637 weakly excites
  24. L13:831 weakly excites

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 -2.32 and -2.171 values between -1.86 and -1.713 values between -1.71 and -1.562 values between -1.56 and -1.414 values between -1.41 and -1.263 values between -1.26 and -1.118 values between -1.11 and -0.967 values between -0.96 and -0.8122 values between -0.81 and -0.6530 values between -0.65 and -0.5061 values between -0.50 and -0.35124 values between -0.35 and -0.20604 values between -0.20 and -0.051952 values between -0.05 and 0.10506 values between 0.10 and 0.26195 values between 0.26 and 0.41131 values between 0.41 and 0.5672 values between 0.56 and 0.7155 values between 0.71 and 0.8629 values between 0.86 and 1.0128 values between 1.01 and 1.1615 values between 1.16 and 1.3111 values between 1.31 and 1.477 values between 1.47 and 1.626 values between 1.62 and 1.773 values between 1.77 and 1.927 values between 1.92 and 2.075 values between 2.07 and 2.222 values between 2.22 and 2.374 values between 2.37 and 2.532 values between 2.53 and 2.682 values between 2.68 and 2.831 values between 2.83 and 2.982 values between 2.98 and 3.131 values between 3.13 and 3.281 values between 3.28 and 3.431 values between 3.59 and 3.74offnegativepositiveoften

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. GED
  2. dreaded
  3. <decltype
  4. gaping
  5. 方方面面
  6. ozilla
  7. 等内容
  8. このような
  9. figcaption
  10. こんな

Suppresses

  1. increment
  2. قيد
  3. ion
  4. ión
  5. l
  6. 束
  7. Increment
  8. rement
  9. IONS
  10. -ne