Neuron Explorer

Qwen3-0.6B

'into' in 'take into account'

A neuron in layer 22 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 responds to the word 'into' in the phrase 'take into account' and similar phrases like 'taking into account'. When it was turned up, the model kept writing the word 'into' again and again.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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

    plant species and 76 percent of alcoholic extracts from the plants contained elevated toxicity levels.⏎The results reinforce the need for traditional preparation methods to take different toxicity levels into account when choosing the appropriate solvent for the preparation of a medicinal remedy. The study

  2. #2 peak

    the mid-point of that 3-year period, with adjustments for incomplete identification by Indigenous status.⏎The Australian headline life expectancy estimates are calculated taking age-specific identification rates into account.⏎The Australian total includes all states and territories (including Other Territories).⏎The

  3. #3 peak

    index of over 1,000 men and women. The researchers used BMI as an indicator of whether the participants were overweight or obese. Even when taking into account other factors, like exercise and total calories consumed, people who ate chocolate regularly

  4. #4 peak

    or information to assess the borrower's financial stability. In this case, the lender may use an average of the borrower's income over the past few years or take into account the borrower's other sources of income to determine their ability to repay the loan

  5. #5 peak

    . Women are generally considered to be paid less than men. There are two distinct numbers regarding the pay gap: unadjusted versus adjusted pay gap. The latter takes into account factors that contribute to the gap, such as differences in education, job experience

  6. #6 peak

    became available nationwide, and already more than half of all eligible schools have elected to participate. The decision to implement CEP is made by the school district, taking into

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 have a 35333333333333333333333333333333333333333333

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

    ’t until 1775 that Alexander Cummings added the u-bend and then 1858 and the Great Stink in London when things really took off. In response to the foul sewage problems of the day, the Government commissioned

  2. #2 peak

    of the more than 900 men aboard, only some 230 survived.⏎In 1953, the Batepa Massacre took place in Sao Tome as Portuguese troops killed some 1,000 striking

  3. #3 peak

    . For example if we are a part of a culture that values ‘modesty’, we may fall into blaming the victim if we deem her behaviour before the crime took place as ‘immodest’.⏎Despite

  4. #4 peak

    collaborating with other data scientists. When you sign up, you get free access to Watson Studio and all other IBM services for 30 days. Start now and take advantage of this offer.⏎- Learn how it performs at speeds up to 1

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. L19:2882 excites
  2. L21:1640 suppresses
  3. L21:1271 suppresses
  4. L20:2454 suppresses
  5. L21:594 weakly excites
  6. L19:2707 weakly excites
  7. L20:512 weakly excites
  8. L14:1303 weakly excites
  9. L21:1538 weakly excites
  10. L9:2404 weakly suppresses
  11. L21:2246 weakly suppresses
  12. L4:2984 weakly excites
  13. L20:107 weakly suppresses
  14. L17:561 weakly suppresses
  15. L12:433 weakly suppresses
  16. L14:257 weakly suppresses
  17. L3:1861 weakly suppresses
  18. L9:1070 weakly excites
  19. L11:2902 weakly suppresses
  20. L9:378 weakly excites
  21. L0:246 weakly excites
  22. L7:684 weakly suppresses
  23. L3:2821 weakly excites
  24. L20:2186 weakly suppresses

Feeds later layers

  1. L27:1250 strongly excites
  2. L27:1849 excites
  3. L27:1564 excites
  4. L24:1333 excites
  5. L26:1582 excites
  6. L23:2056 excites
  7. L23:810 excites
  8. L23:1182 excites
  9. L23:1359 excites
  10. L26:1526 excites
  11. L25:1838 weakly excites
  12. L25:1153 weakly suppresses
  13. L25:2479 weakly excites
  14. L25:1114 weakly excites
  15. L24:2721 weakly suppresses
  16. L26:2162 weakly suppresses
  17. L23:1396 weakly excites
  18. L23:798 weakly suppresses
  19. L27:2748 weakly excites
  20. L24:184 weakly excites
  21. L24:2662 weakly suppresses
  22. L26:2702 weakly suppresses
  23. L26:1355 weakly excites
  24. L24:1022 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 -13.79 and -11.951 values between -8.27 and -6.432 values between -6.43 and -4.592 values between -4.59 and -2.7527 values between -2.75 and -0.913812 values between -0.91 and 0.9341 values between 0.93 and 2.775 values between 2.77 and 4.615 values between 4.61 and 6.454 values between 6.45 and 8.292 values between 8.29 and 10.132 values between 10.13 and 11.971 values between 15.65 and 17.491 values between 23.01 and 24.851 values between 26.69 and 28.531 values between 57.97 and 59.81offnegativepositiveoften

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. take
  2. takes
  3. taking
  4. took
  5. take
  6. taking
  7. Take
  8. Taking
  9. Take
  10. Taking

Suppresses

  1. thur
  2. 揭秘
  3. stype
  4. $/
  5. .addButton
  6. CreateTable
  7. _EXCEPTION
  8. esen
  9. umpt
  10. messageType