Neuron Explorer

Qwen3-0.6B

'long' in 'how long' questions

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 word 'long' in questions like 'how long will it take'. Its opposite side reacts to the word after 'How' in other questions, such as 'how is' or 'how are'. Turning it up did not visibly change the writing.

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

    how long they will be sick, exposed to dialysis or ventilators. The less time

  2. #2 peak

    line at the point Q. Prove AQ = AD.⏎At the time of writing the hour and minute hands of my clock are at⏎right angles. How long will it be before they

  3. #3 peak

    ⏎Find out what he or she uses⏎the computer for⏎The time spent at the computer⏎Keep track of how long your child⏎is on the computer⏎How Visikid works?⏎This small

  4. #4 peak

    el per litre adult limit in parts of Ukraine until at least 2040.⏎“This situation should also act as a warning and a reminder of just how long the legacy of nuclear accidents can be,” said Labuns<|endoftext|>Gun assaults in school

  5. #5 peak

    ilers seek to push their gas tanks to the limit and achieve hypermileage, exceeding EPA ratings for miles per gallon.⏎So, there you have it. How long 'till hyper is replaced by wondermileage or excessivemileage

  6. #6 peak

    aquarium spokesperson, said alligator snapping turtles are believed to live between 80 to 120 years in the wild. It currently isn't known how long they can live under professional care, she said, because zoos and aquariums

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 have been in the past, and I have been in the past, and I have been in the past. So I think of this, and I have been in the past, and I have been in the past. So I think

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

    fracking pollute the water supply? Will fracking make the United States energy independent? Does fracking cause earthquakes? How is fracking regulated? Is fracking good for the economy? Coupling a deep understanding of

  2. #2 peak

    how are these numbers obtained? And wouldn’t everything feature UPF50+?⏎There

  3. #3 peak

    a great addition to Lesson 1 – Sustainable Food and Agriculture⏎What are GMOs and are they okay to eat? What does organic farming look like? How are animals like chickens or cattle raised?⏎snapAG is a series of resources that invite

  4. #4 peak

    just ignore those rules, any more⏎Continue reading →<|endoftext|>How are ayurvedic treatments effective?⏎Originating in the Indian subcontinent over 5

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. L2:721 excites
  2. L1:124 weakly suppresses
  3. L15:509 weakly excites
  4. L1:903 weakly suppresses
  5. L13:2936 weakly excites
  6. L20:1759 weakly excites
  7. L11:824 weakly suppresses
  8. L16:328 weakly excites
  9. L9:264 weakly suppresses
  10. L15:1945 weakly suppresses
  11. L25:33 weakly excites spot where a number is due
  12. L19:1310 weakly suppresses
  13. L9:2580 weakly suppresses
  14. L18:2213 weakly suppresses
  15. L8:2075 weakly excites
  16. L3:351 weakly excites
  17. L1:189 weakly suppresses
  18. L4:3011 weakly suppresses
  19. L4:1729 weakly suppresses
  20. L22:945 weakly suppresses
  21. L26:318 weakly suppresses
  22. L25:3032 weakly suppresses 'therm' in science words
  23. L3:661 weakly suppresses
  24. L4:154 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 -2.24 and -2.081 values between -1.93 and -1.781 values between -1.62 and -1.471 values between -1.16 and -1.0112 values between -0.55 and -0.3940 values between -0.39 and -0.24245 values between -0.24 and -0.092840 values between -0.09 and 0.07745 values between 0.07 and 0.224 values between 0.22 and 0.382 values between 0.38 and 0.532 values between 0.53 and 0.681 values between 0.68 and 0.841 values between 0.84 and 0.992 values between 0.99 and 1.151 values between 1.30 and 1.451 values between 1.45 and 1.612 values between 1.61 and 1.761 values between 1.76 and 1.921 values between 2.22 and 2.382 values between 3.61 and 3.762 values between 3.76 and 3.91offnegativepositiveoften

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. .Cast
  2. antage
  3. $MESS
  4. 竹
  5. *)_
  6. inkel
  7. ivalence
  8. CAST
  9. erte
  10. be

Suppresses

  1. arma
  2. [color
  3. 启示
  4. 专区
  5. อาร
  6. 梯队
  7. squads
  8. ourg
  9. PJ
  10. inse