Neuron Explorer

Qwen3-0.6B

start of a new document

A neuron in layer 0 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 separator the training text uses between web pages, specifically on the first word after that separator. It also turns slightly negative on the word 'fructose'. Steering made the output repetitive without a clear link to the trigger.

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

    your own microclimate," architect and Natural Home columnist Carol Venolia says. "Working with nature is simple, but we lost touch with that ability when we became<|endoftext|>THE NEED FOR CULTURAL REVOLUTION⏎Prof. Jose Ma. Sison

  2. #2 peak

    smartest thing you ever did.Powered by Sidelines<|endoftext|>There is a wealth of scientific literature available on the treatment of adults diagnosed with schizophrenia

  3. #3 peak

    <|endoftext|>House in Space, from a NASA Remix Challenge. Credit: Cookieater20

  4. #4 peak

    <|endoftext|>10 Tips for Parents of Prospective College Students⏎1. Choosing a Career

  5. #5 peak

    <|endoftext|>Active transportation refers to non-motorized transportation modes, such as bicycling and

  6. #6 peak

    <|endoftext|>Public holidays in Pakistan are different than bank holidays in the country. While all of

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 current ) as function 1 1. What system What What What topics are: the question: questions: What titles. The questions you will questions. What titles: questions.

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

    fructose-1-6⏎diphosphate deficiency and in fructose-1-phosphate aldolase deficiency⏎(hereditary fructose intolerance). Fructose-1-6 diphosphate deficiency⏎produces accumulation of lactic and

  2. #2 peak

    diet. Continuous vomiting leads to coma. Carbohydrate abnormalities are⏎diagnosed by finding reduced enzyme activity in the liver. Treatment consists⏎of eliminating fructose from the diet and correcting hypoglycemia.⏎syndrome is produced

  3. #3 peak

    and potatoes are classified as complex carbohydrates. However, the body converts starches to blood sugar as rapidly as it processes pure glucose. On the other hand, fructose (fruit sugar) is branded as a simple carbohydrate, yet has a minimal effect

  4. #4 peak

    , so it must first sever the chemical link connecting the two sugars. The enzyme sucrase in your small intestine assists with the breakdown of sucrose into fructose and glucose. This allows your body to absorb them, transport them to the liver

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

None: this is the first layer.

Feeds later layers

  1. L1:25 excites
  2. L1:2 excites
  3. L3:24 excites
  4. L1:7 excites
  5. L3:1256 excites
  6. L1:57 excites
  7. L1:109 excites
  8. L1:49 excites
  9. L3:1 suppresses
  10. L1:2554 excites
  11. L5:931 suppresses
  12. L4:11 excites
  13. L1:1227 suppresses
  14. L3:8 excites
  15. L1:42 suppresses
  16. L16:769 excites
  17. L1:2618 excites
  18. L3:504 excites
  19. L2:52 excites
  20. L1:751 suppresses
  21. L2:2128 excites
  22. L1:2012 excites
  23. L2:47 excites
  24. L5:574 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 -0.57 and -0.531 values between -0.53 and -0.482 values between -0.48 and -0.434 values between -0.43 and -0.389 values between -0.38 and -0.3419 values between -0.34 and -0.2933 values between -0.29 and -0.2466 values between -0.24 and -0.19119 values between -0.19 and -0.15293 values between -0.15 and -0.10629 values between -0.10 and -0.051055 values between -0.05 and -0.001040 values between -0.00 and 0.05465 values between 0.05 and 0.09107 values between 0.09 and 0.1436 values between 0.14 and 0.1912 values between 0.19 and 0.243 values between 0.24 and 0.282 values between 0.28 and 0.332 values between 0.33 and 0.383 values between 1.00 and 1.051 values between 1.05 and 1.092 values between 1.09 and 1.142 values between 1.14 and 1.191 values between 1.28 and 1.33offnegativepositiveoften

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. -------------</
  2. 第二个
  3. 什么
  4. ROUND
  5. THON
  6. омн
  7. жен
  8. PLIC
  9. *</
  10. ;'⏎

Suppresses

  1. ettes
  2. 商报
  3. 鳅
  4. like
  5. nowadays
  6. checker
  7. blick
  8. lette
  9. iever
  10. plus