Neuron Explorer

Qwen3-0.6B

"well" in "as well as"

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 lights up on the word "well" in the phrase "as well as", where something is being added next to something else. Turning it up or down did not make the model write about that phrase, so the link to the output is unclear.

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

    circuits that can be sprayed onto anything using modified inkjet technologies - no expensive clean rooms necessary. That has the attention of dozens of fledgling businesses, as well as industry veterans Dow, IBM, Philips, and Xerox, which are pouring

  2. #2 peak

    ships so that they could beat competitors and outwit planetary security forces. Almost every smugglers vessel had improved light speed and sub-light speed drives for increased speed as well as boosted weapons systems to fight their way out of tough situations. Another of the

  3. #3 peak

    types of restoration strategies will be used for this project. At least 50,000 native trees and shrubs will be planted during the project as well as several thousand wildflowers and grasses in order to restore approximately 25

  4. #4 peak

    , including interviews and reporting on the maker market and the wider technology world. Our standards page is intended as a guide to best practices that Adafruit uses, as well as an outline of the ethical standards Adafruit aspires to. While Adafruit

  5. #5 peak

    concessions. If history tells us anything, his turn toward a more populist brand of politics has more to do with the energy of reform movements within Honduras itself, as well as throughout the rest of Latin America, than any personal awakening.⏎From the shift

  6. #6 peak

    with the precise soundbites that will accompany them which match what was actually recorded in interviews⏎Let’s jump in and examine each of these in detail, as well as learn how to write each one.⏎What is a documentary film treatment?⏎A

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 AM A TALKING FISH, AND I AM A WALKING FISH, AND I AM A FISH BUT I AM A FISH, AND I AM A FISH, AND I AM A FISH, AND I

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

    9/10)k-2, where k is the number of digits of 2n. The number of digits can be estimated as n log102. Thus, the probability is about cxn, where c = (1

  2. #2 peak

    Servicing Ratio is a measure used by lenders to assess a borrower's ability to repay their debts. The DSR is calculated by dividing a borrower's monthly income by their monthly debt repayments, including mortgage payments, credit card payments, and other

  3. #3 peak

    .⏎Hence, the given lines are parallel.⏎Let's consider another set of equations.⏎y = 2x +3 and 4y = –2x+10⏎On comparing y = 2x +3 and 4

  4. #4 peak

    to identify parallel and perpendicular lines.⏎Let's take some examples to understand this.⏎Considering two equations:⏎y = 2x +3 and y = 2x+5⏎On comparing y = 2x +3 and y =

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. L12:2659 suppresses
  2. L14:2732 excites
  3. L13:1312 suppresses
  4. L14:2020 suppresses
  5. L12:2696 suppresses
  6. L14:2768 suppresses
  7. L14:2943 weakly suppresses
  8. L5:108 weakly suppresses
  9. L9:2939 weakly suppresses
  10. L12:2058 weakly excites
  11. L14:986 weakly suppresses
  12. L11:2336 weakly suppresses
  13. L8:1069 weakly suppresses
  14. L14:1361 weakly excites
  15. L8:711 weakly excites
  16. L13:781 weakly suppresses
  17. L10:1205 weakly excites
  18. L0:850 weakly suppresses
  19. L14:3004 weakly excites
  20. L11:1655 weakly suppresses
  21. L4:1788 weakly suppresses
  22. L14:2880 weakly suppresses
  23. L7:660 weakly excites
  24. L13:2824 weakly excites

Feeds later layers

  1. L16:1084 suppresses
  2. L16:473 excites
  3. L20:1763 suppresses
  4. L16:3060 suppresses
  5. L16:2880 suppresses
  6. L16:2386 suppresses
  7. L17:2450 weakly suppresses
  8. L16:2329 weakly suppresses
  9. L18:2331 weakly suppresses
  10. L19:2800 weakly suppresses
  11. L18:1573 weakly excites
  12. L16:533 weakly excites
  13. L17:2727 weakly excites
  14. L21:2925 weakly excites
  15. L19:2225 weakly suppresses
  16. L16:774 weakly suppresses
  17. L16:1956 weakly excites
  18. L25:2200 weakly excites
  19. L19:1262 weakly suppresses
  20. L17:1488 weakly suppresses
  21. L21:1338 weakly excites
  22. L16:2790 weakly excites
  23. L17:632 weakly suppresses unclear: coordinating words in lists
  24. L24:2510 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.

2 values between -1.27 and -1.142 values between -0.65 and -0.5216 values between -0.52 and -0.4075 values between -0.40 and -0.28324 values between -0.28 and -0.15905 values between -0.15 and -0.031391 values between -0.03 and 0.10796 values between 0.10 and 0.22301 values between 0.22 and 0.3477 values between 0.34 and 0.4712 values between 0.47 and 0.592 values between 0.59 and 0.712 values between 0.71 and 0.841 values between 0.84 and 0.961 values between 2.08 and 2.201 values between 3.56 and 3.69offnegativepositiveoften

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. oft
  2. 有效的
  3. chos
  4. necess
  5. nas
  6. 更好的
  7. monkey
  8. ammer
  9. 铤
  10. estará

Suppresses

  1. enberg
  2. 然后
  3. изма
  4. tical
  5. CREMENT
  6. 并
  7. 但是我
  8. IMIZE
  9. 年以上
  10. 旋