Neuron Explorer

Qwen3-0.6B

'off of' and 'of' after 'off'

A neuron in layer 2 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 "of" when it comes right after "off", as in "feed off of blood" or "off of Broadway". Turning it up did not produce a clear shift in the model's 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

    as clothes, carpets, and, yes, beds, and they feed off of blood.⏎There are about ninety known species of bed bugs worldwide, but luckily for

  2. #2 peak

    Manitowoc Rapids. It is estimated that more than 5,000 people attended the event. Today, the marker is located off of Broadway Street.⏎In the days before settlement, our area was named Manitowoc

  3. #3 peak

    the world’s most fragile and endangered ecosystems. In the last few decades over 35 million acres of Coral Reefs have been obliterated. Reefs off of <|endoftext|>Time Ticks Away in Portugal⏎Compared with every other country in

  4. #4 peak

    red squirrels and porcupines.⏎Types of Flora⏎The animals of the park have plenty to forage and thrive off of with a strong botanical environment in Palo Verde National Park. Some of the plant species

  5. #5 peak

    using simple methods.⏎Now, caffeine has fairly good solubility in a variety of environments thanks to all those myriad polar and non-polar functional groups hanging off of it, which results in a wide range of possibilities for means of extraction, collection

  6. #6 peak

    group activities, toys, and other things the children will do within a week. Creating a newsletter for parents. I observed children and their development and assessed them off of what I observed. I also observed planning for field trips. I interacted with

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 working on some problems. I have been, I have been thinking about. experiences. the projects. this is. . The. The key, I can. will. is. online. will. activities in. course.

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: Differentiate between www and internet.⏎problem 10: What do you mean by Usenet? List out main categories of it?⏎problem 11: Define

  2. #2 peak

    down, but cutting would only be done when the beetle is dormant to prevent further spreading, Eddy says. She adds, "There is no one textbook way of dealing with this because there can be so many different circumstances."⏎For now, the

  3. #3 peak

    methods you can use to overcome the blue light effects without cutting your TV watch time.⏎1. Protective Glasses:⏎Using protective glasses is one of the most effective ways of overcoming the problems of blue light. The glasses are available in the market that effectively

  4. #4 peak

    most successful people share the common trait of self-awareness. They recognize the situations that will make them successful, and this makes it easy for them to find ways of achieving objectives that resonate with their motivations.⏎They also understand their limitations and where they

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. L1:2234 weakly suppresses
  2. L1:1129 weakly suppresses
  3. L1:1579 weakly suppresses 'Well' starting a reply or answer
  4. L1:2559 weakly excites
  5. L1:2945 weakly excites
  6. L0:1898 weakly suppresses
  7. L1:669 weakly excites
  8. L0:786 weakly suppresses
  9. L0:274 weakly excites
  10. L1:1152 weakly excites
  11. L1:1876 weakly suppresses
  12. L1:2268 weakly excites
  13. L1:535 weakly suppresses
  14. L1:2187 weakly excites
  15. L0:1651 weakly suppresses
  16. L1:1097 weakly excites
  17. L0:2093 weakly suppresses
  18. L0:1580 weakly excites
  19. L0:1499 weakly excites
  20. L1:1308 weakly excites
  21. L1:2620 weakly excites
  22. L1:149 weakly suppresses
  23. L1:2465 weakly excites
  24. L1:924 weakly excites

Feeds later layers

  1. L9:1782 suppresses
  2. L7:1157 excites
  3. L17:664 suppresses
  4. L8:1251 suppresses
  5. L8:514 suppresses
  6. L3:1525 suppresses
  7. L9:2913 excites
  8. L10:914 excites
  9. L10:664 suppresses
  10. L24:456 excites
  11. L24:1748 weakly suppresses
  12. L10:1042 weakly excites
  13. L14:979 weakly suppresses
  14. L9:1625 weakly suppresses
  15. L7:437 weakly suppresses
  16. L8:277 weakly excites
  17. L10:190 weakly suppresses
  18. L20:1637 weakly suppresses
  19. L26:843 weakly excites
  20. L16:692 weakly suppresses
  21. L7:887 weakly excites
  22. L24:824 weakly suppresses
  23. L23:3035 weakly suppresses
  24. L3:660 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 -1.67 and -1.531 values between -1.53 and -1.402 values between -1.13 and -0.996 values between -0.99 and -0.864 values between -0.86 and -0.729 values between -0.72 and -0.5912 values between -0.59 and -0.4530 values between -0.45 and -0.32120 values between -0.32 and -0.18681 values between -0.18 and -0.052487 values between -0.05 and 0.09420 values between 0.09 and 0.2292 values between 0.22 and 0.3624 values between 0.36 and 0.497 values between 0.49 and 0.622 values between 0.62 and 0.762 values between 0.76 and 0.893 values between 0.89 and 1.031 values between 1.43 and 1.571 values between 1.70 and 1.841 values between 1.97 and 2.111 values between 2.11 and 2.241 values between 3.59 and 3.73offnegativepositiveoften

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. -san
  2. erto
  3. [dir
  4. -driving
  5. cimal
  6. atron
  7. stered
  8. irement
  9. 发改委
  10. 指南

Suppresses

  1. /Error
  2. igham
  3. onga
  4. ึง
  5. /Core
  6. taş
  7. 岙
  8. ymce
  9. PAS
  10. og