Neuron Explorer

Qwen3-0.6B

dots inside version and section numbers

A neuron in layer 11 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 dot inside numbers like version '0.5.0', section '2.2', or IP-style '188.8.131.52'. Turning it up made the model produce strings of '1.1.1.1' patterns, which fits the same idea.

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

    1 by Michael Faraday, and share several key functional parts.⏎- 1 Power transformers⏎- 1.1 Laminated core⏎- 1.2 Toroidal⏎- 1.3 Autotransformer⏎-

  2. #2 peak

    for-mation of the dissimilar metal joint and the measurement of joint tensile strength⏎Welding Method 6, Paragraph 188.8.131.52 (Paraphrased, still in public comment

  3. #3 peak

    grow a new Monstera plant effectively:-⏎- 2.1 Step 1: Finding the right place to grow a new plant:⏎- 2.2 Step 2: Search for the right node or petiole:⏎-

  4. #4 peak

    <|endoftext|>Mercury is used in thermometers, fluorescent light bulbs, and some electrical switches.⏎"Mercury" is in the HS Code "2805.40"⏎- - Mercury⏎

  5. #5 peak

    Figure 3-1).⏎Extension modules play an active role in all these aspects of the Apache server's⏎Figure 3-1. The HTTP transaction consists⏎of a URI request from the browser to

  6. #6 peak

    the hood. If the face velocity becomes excessive, the facility engineers should make adjustments or corrections.⏎For hoods without face velocity controls (see section 8.C.6.

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," you say, "What is the meaning of 'not work for you'?" So I'm wondering, what is the meaning of 'I'?" in the context of the word 'today'?" So 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

    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

  2. #2 peak

    to call these ‘selfish’ elements ‘gene drives’ is not accurate. Though they are to an extent based on naturally occurring systems, CRISPR-Cas9 and/or other engineered gene drives are very different than these naturally occurring selfish genes

  3. #3 peak

    1: Download Drupal⏎First download the Drupal using following command:⏎# wget http://ftp.drupal.org/files/projects/drupal-7.33.zip⏎Step 2: Extract Compressed Drupal⏎

  4. #4 peak

    means that the files are stored in a compressed format.⏎For Python Implementation⏎You can use OpenCV to manipulate the images.⏎pip3 install opencv-python numpy⏎Open up a new Python file and follow along:⏎import numpy as np

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. L8:1107 excites
  2. L10:2461 excites
  3. L5:475 weakly excites
  4. L3:1831 weakly suppresses
  5. L5:2928 weakly excites
  6. L3:1427 weakly excites
  7. L10:3060 weakly suppresses
  8. L6:1087 weakly suppresses
  9. L3:723 weakly excites
  10. L9:1587 weakly excites
  11. L5:885 weakly excites
  12. L5:3016 weakly suppresses
  13. L8:1618 weakly suppresses
  14. L9:2503 weakly excites
  15. L7:776 weakly suppresses
  16. L10:623 weakly excites
  17. L8:1672 weakly excites
  18. L10:1516 weakly excites
  19. L10:1510 weakly excites
  20. L9:1658 weakly excites
  21. L6:579 weakly excites
  22. L9:95 weakly suppresses
  23. L0:66 weakly suppresses
  24. L7:552 weakly suppresses

Feeds later layers

  1. L16:372 excites
  2. L19:298 excites
  3. L14:1265 excites
  4. L13:1083 excites
  5. L21:395 excites
  6. L13:2041 excites
  7. L12:2379 excites period or digit inside section numbers
  8. L17:1806 excites
  9. L13:1426 excites
  10. L17:870 excites
  11. L20:2450 weakly suppresses
  12. L14:2346 weakly excites
  13. L13:2749 weakly excites
  14. L13:727 weakly excites
  15. L14:2044 weakly suppresses
  16. L16:1800 weakly excites
  17. L15:1255 weakly excites
  18. L22:2931 weakly excites
  19. L24:2389 weakly excites
  20. L23:1436 weakly excites
  21. L19:728 weakly excites
  22. L26:316 weakly excites
  23. L13:1772 weakly excites
  24. L21:175 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 -1.09 and -1.011 values between -0.78 and -0.701 values between -0.70 and -0.6213 values between -0.62 and -0.5415 values between -0.54 and -0.4623 values between -0.46 and -0.3955 values between -0.39 and -0.31101 values between -0.31 and -0.23261 values between -0.23 and -0.15550 values between -0.15 and -0.07904 values between -0.07 and 0.01943 values between 0.01 and 0.09570 values between 0.09 and 0.16262 values between 0.16 and 0.24125 values between 0.24 and 0.3247 values between 0.32 and 0.4017 values between 0.40 and 0.486 values between 0.48 and 0.563 values between 0.56 and 0.632 values between 0.63 and 0.711 values between 0.79 and 0.871 values between 0.95 and 1.031 values between 1.26 and 1.342 values between 1.34 and 1.421 values between 1.58 and 1.651 values between 1.81 and 1.891 values between 1.97 and 2.05offnegativepositiveoften

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. icles
  2. Ivanka
  3. tep
  4. nonetheless
  5. appendix
  6. icle
  7. acknow
  8. subsection
  9. enda
  10. ément

Suppresses

  1. ere
  2. طلاق
  3. ps
  4. pee
  5. ím
  6. 负
  7. ze
  8. igest
  9. public
  10. ill