Neuron Explorer

Qwen3-0.6B

unclear: scattered code and URL tokens

A neuron in layer 6 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron is hard to pin down. Its positive responses are weak and fall on scattered pieces of URLs, software names and technical words, with no shared meaning. Its strongest negative responses are on words like 'function' in code and on a comma after a copyright year. Turning it up gave unrelated text.

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

    population growth or technical progress. Production per capita, at date t, is given by⏎where k denotes capital per capita and A is a productivity parameter. Capital depreciates at the rate dt, where dt is a convex function

  2. #2 peak

    of powerful, dynamic applications and services. For more information on this great topic, please consult:⏎- Sun's Java Server Pages home (http://java.sun.com/products/jsp/)⏎- Westy Rockwell's article on "JSP Tag Libraries

  3. #3 peak

    EDIS (keyword search "marketing") and on the UF/IFAS Tropical Research and Education Center (TREC) Agricultural Economics Extension website (http://agecon.centers.ufl.edu/) (Evans and Ballen 201

  4. #4 peak

    http://www.tandfonline.com/pricing/journal/VECE20|⏎When requesting a correction, please mention this item's handle: RePEc:taf:jeduce:v:42:y:2011

  5. #5 peak

    specific section of an organism’s genome, causing the organism to behave differently and to pass this difference to all their offspring. CRISPR-Cas9 gene drives do this cut-copy-paste during the reproductive cycle to make sure

  6. #6 peak

    in regular classes.⏎If you experience problems downloading a file, check if you have the proper application to view it first. In case of further problems read the IDEAS help page. Note that these files are not on the IDEAS site. Please

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 using a simple script to implement a function that takes a list of integers and returns their sum. Function: Compute the sum of all integers in the list """ # This script is for a simple task to calculate the sum

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

    . The exact solution can be found by quadrature and is⏎To use MATLAB we first construct an M-file that defines the equation.⏎% filename rhs.m⏎function ydot=rhs(t,y)%compute the rhs of eqns.⏎%for

  2. #2 peak

    © 2012, Columbia University Press. All rights reserved.⏎More on rheology from Fact Monster:⏎

  3. #3 peak

    -rolling and calculating outcomes is handled for the players by the software.⏎-- MarkMusante - 11 Sep 2002⏎Influence on IF⏎Adventure and

  4. #4 peak

    when you have several commands that you will use over and over it is more convenient to create a script that can run all the commands with one message from you.⏎% filename simple.m⏎ini=1.; %set the variable ini 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. L2:322 excites
  2. L3:5 weakly excites
  3. L2:1784 weakly excites
  4. L5:228 weakly suppresses
  5. L5:50 weakly excites
  6. L2:41 weakly suppresses
  7. L0:178 weakly suppresses
  8. L4:2992 weakly excites
  9. L4:1836 weakly excites
  10. L3:63 weakly excites
  11. L0:1224 weakly suppresses
  12. L2:2399 weakly excites
  13. L1:2347 weakly suppresses
  14. L1:1234 weakly suppresses
  15. L4:2254 weakly excites
  16. L4:1689 weakly excites
  17. L1:2683 weakly excites
  18. L0:1301 weakly suppresses
  19. L4:1615 weakly excites
  20. L4:1867 weakly suppresses
  21. L0:1800 weakly suppresses
  22. L3:148 weakly excites
  23. L2:1136 weakly excites
  24. L5:471 weakly suppresses

Feeds later layers

  1. L7:2319 suppresses
  2. L9:114 suppresses
  3. L9:100 suppresses
  4. L8:2922 suppresses
  5. L10:1791 weakly suppresses
  6. L11:2244 weakly suppresses
  7. L8:1361 weakly suppresses
  8. L12:2106 weakly suppresses
  9. L9:642 weakly excites
  10. L13:2300 weakly excites
  11. L10:2794 weakly suppresses
  12. L13:3005 weakly suppresses
  13. L22:1703 weakly suppresses
  14. L7:2430 weakly excites
  15. L7:2491 weakly suppresses
  16. L7:730 weakly suppresses
  17. L27:1254 weakly excites
  18. L8:1187 weakly suppresses
  19. L21:1973 weakly excites
  20. L8:932 weakly excites
  21. L13:1463 weakly excites
  22. L9:2017 weakly suppresses
  23. L12:1805 weakly excites
  24. L8:1729 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.

1 values between -0.94 and -0.901 values between -0.42 and -0.384 values between -0.38 and -0.333 values between -0.33 and -0.2915 values between -0.29 and -0.2517 values between -0.25 and -0.2047 values between -0.20 and -0.1678 values between -0.16 and -0.12172 values between -0.12 and -0.07323 values between -0.07 and -0.03662 values between -0.03 and 0.02742 values between 0.02 and 0.06595 values between 0.06 and 0.10411 values between 0.10 and 0.15287 values between 0.15 and 0.19192 values between 0.19 and 0.23127 values between 0.23 and 0.2885 values between 0.28 and 0.3249 values between 0.32 and 0.3627 values between 0.36 and 0.4128 values between 0.41 and 0.4513 values between 0.45 and 0.499 values between 0.49 and 0.548 values between 0.54 and 0.585 values between 0.58 and 0.624 values between 0.62 and 0.672 values between 0.67 and 0.711 values between 0.75 and 0.80offnegativepositiveoften

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. عد
  2. rel
  3. URE
  4. ности
  5. 要在
  6. 照样
  7. ность
  8. nesia
  9. _second
  10. 行李

Suppresses

  1. '],⏎
  2. '], ⏎
  3. enna
  4. oras
  5. oria
  6. */,⏎
  7. Thou
  8. hot
  9. bó
  10. -ton