Neuron Explorer

Qwen3-0.6B

'to' after a noun, before a purpose verb

A neuron in layer 15 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 'to' when it comes after a noun and tells you what that noun is for, as in 'something to latch on to', 'nothing to talk about' or 'a place to rest'. Turning it up made the model repeat 'to watch' and 'to feel' over and over.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. Think it is wrong? Flag it (opens an email to the maker with the neuron already filled in).

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

    after the larvae are spawned, they reach a period of metamorphosis and go to the bottom of the tank to attach themselves to shell fragments so they have something to latch on to as they begin to grow their own shells. The farmer said any

  2. #2 peak

    atypical from the others because of that. In a way if you don’t share your memories, people will find you boring and dull, because you have nothing to talk about, because everything you talk about is a memory. A person cannot talk

  3. #3 peak

    is full of water but capped off on both ends.⏎You hit the power switch on the pumps, what happens?⏎The answer is nothing, there is no where to move the water to, the pumps don't even spin. (ignore water turbulence

  4. #4 peak

    in the well-being of the surrounding environment. Pollinators are in trouble due to habitat loss, pesticide use and the increasing threat of climate change. With nowhere to build their homes and no plants to pollinate, honeybee

  5. #5 peak

    school is a place to learn and that learning is COOL! Constantly point out things they have learned and things they have yet to learn to give them something to look forward to. Feel free to talk about your own learning experiences with them.⏎

  6. #6 peak

    the hospital⏎To provide these carers with a place to rest while staying at the hospital to look after their sick relatives, the Foundation helped

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 fact that, if there are not enough resources (i) the time and financial constraints (i) the time and financial constraints, and (i) the time and financial constraints) (i) the time and financial constraints) and (

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

    well as calculating summary statistics such as means, frequency, and percentiles. Creating filters and joins on in-memory data are also discussed.Learn how to⏎Who should attendExperienced programmers who need to learn the syntax and functionality of the SAS LAS

  2. #2 peak

    Dosing InformationThe dose of Monopril your healthcare provider recommends will vary depending on a number of factors, including:⏎- Your age⏎- The medical condition being treated⏎- Other medical conditions you may have⏎- Other medications you may currently

  3. #3 peak

    the decimal and cents to the right of the decimal. To add decimals, simply line up the decimal points of the amounts of money that are being added. When you have added the amounts of money, be sure to place the

  4. #4 peak

    help maintain the nation's natural gas transmission infrastructure through the timely and effective detection of natural gas leaks through evaluation of geobotanical stress signatures.⏎The remote sensing techniques being developed employ advanced spectrometer systems that produce visible and near infrared

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. L9:2312 excites 'to' after a noun (a role to play)
  2. L7:1974 excites
  3. L12:2519 suppresses
  4. L6:2297 excites
  5. L7:2227 excites
  6. L14:2265 excites
  7. L11:1811 suppresses
  8. L14:2416 suppresses
  9. L13:1686 suppresses
  10. L14:1854 excites
  11. L4:470 suppresses
  12. L6:958 suppresses
  13. L5:1030 excites
  14. L14:2791 excites
  15. L13:134 excites
  16. L13:2057 suppresses
  17. L12:1542 weakly suppresses
  18. L14:492 weakly suppresses
  19. L6:2182 weakly excites
  20. L12:1084 weakly suppresses
  21. L5:439 weakly suppresses
  22. L7:2553 weakly suppresses
  23. L1:464 weakly excites
  24. L5:1428 weakly suppresses

Feeds later layers

  1. L16:2537 excites
  2. L17:2067 excites
  3. L16:2555 excites
  4. L19:1235 excites
  5. L19:2757 excites
  6. L18:2703 excites
  7. L18:2440 excites
  8. L19:2010 excites
  9. L21:2805 excites 'the' right before 'subject' or 'source'
  10. L18:2666 excites
  11. L23:1542 excites
  12. L16:1700 excites
  13. L18:1672 excites
  14. L17:2575 excites
  15. L19:2621 excites
  16. L18:2740 excites
  17. L21:2602 excites
  18. L22:2402 excites
  19. L17:1884 excites
  20. L22:1422 excites
  21. L17:2756 excites
  22. L16:2474 excites
  23. L22:429 excites
  24. L16:2214 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 -5.93 and -5.691 values between -5.44 and -5.201 values between -4.71 and -4.471 values between -3.74 and -3.491 values between -2.76 and -2.522 values between -2.52 and -2.282 values between -2.03 and -1.793 values between -1.79 and -1.543 values between -1.30 and -1.064 values between -1.06 and -0.8131 values between -0.81 and -0.57266 values between -0.57 and -0.331296 values between -0.33 and -0.081822 values between -0.08 and 0.16400 values between 0.16 and 0.4144 values between 0.41 and 0.6511 values between 0.65 and 0.895 values between 0.89 and 1.144 values between 1.14 and 1.381 values between 1.62 and 1.873 values between 1.87 and 2.111 values between 2.11 and 2.351 values between 2.35 and 2.601 values between 2.60 and 2.842 values between 3.09 and 3.331 values between 3.57 and 3.82offnegativepositiveoften

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. 可供
  2. to
  3. holds
  4. 可
  5. which
  6. hold
  7. ried
  8. otted
  9. to
  10. which

Suppresses

  1. alley
  2. 家伙
  3. ouce
  4. SharePoint
  5. iences
  6. 噜
  7. uem
  8. SystemService
  9. 缯
  10. <num