Neuron Explorer

Qwen3-0.6B

'on' in 'went on to'

A neuron in layer 5 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

The neuron fires on the word 'on' in phrases like 'went on to' or 'go on to', where someone later does something. Turning it up did not clearly change the writing style.

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

    and small, in the sciences as well as in the humanities.⏎Attending a lecture and taking notes works very well for some students, particularly the students who go on to careers in academia. They’re able to assimilate and process the information shared

  2. #2 peak

    to Alzheimer’s in roughly 10% to 15% of cases each year, according to the study.⏎Forty-five percent of the participants went on to develop serious dementia consistent with Alzheimer’s disease. Those who had high cholesterol,

  3. #3 peak

    and Technology (NIST) and replaced by the Advanced Encryption Standard (AES), Diebold went on to hardcode it into the source code—a willful circumvention of best practices

  4. #4 peak

    This mansion in the Vine neighborhood, built in 1910, was the home of Alonzo Herndon, who was born into slavery but went on to become Atlanta's first black millionaire. (Photo Courtesy by Philip Graitcer

  5. #5 peak

    Schultz, PhD, director of the Center for Autism Research at CHOP and a co-author on the study.⏎These findings suggest that 7-month-olds who go on to develop autism show subtle, yet overt, behavioral differences prior to the emergence of

  6. #6 peak

    action for equality. The first IWD was acknowledged in the U.S. in 1909 when 15,000 female workers went on strike in New York City to demand improved working conditions, shorter working hours and voting

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:** ##** A** All ` ##** A** A** A** A** A** A** A** A** A** A** A** A** A** A** A** A** A** A** A**

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

    times the mass of the sun was converted to energy in a fraction of a second.⏎The discovery has been accepted for publication in Physical Review Letters.⏎Gravitational waves are ripples in the universe caused by some of the most energetic cosmic events,

  2. #2 peak

    ipples started to get serious.⏎But the first generation LIGO experiment, which ran for eight years, wasn’t sensitive enough. Which is understandable. Gravitational waves are minuscule — the atomic jitters that pass through our world when two

  3. #3 peak

    modo have been flying for a month, the LIGO team wanted to be absolutely certain before making an official announcement.⏎That announcement has just come. Gravitational waves were observed on 14 September 2015, at 5

  4. #4 peak

    a century ago, physicists have been on the hunt for gravitational waves, ripples in the fabric of spacetime. That hunt is now over. Gravitational waves exist, and we’ve found them.⏎That’s according to researchers at the Laser

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. L3:2935 weakly suppresses
  2. L4:1724 weakly suppresses
  3. L0:1085 weakly excites
  4. L1:2313 weakly suppresses
  5. L4:2082 weakly suppresses
  6. L0:1572 weakly excites
  7. L4:1684 weakly suppresses
  8. L4:1566 weakly excites
  9. L4:2477 weakly suppresses sorting and ordering words
  10. L3:3039 weakly suppresses
  11. L4:1602 weakly excites
  12. L4:2567 weakly suppresses the word 'still' mid-sentence
  13. L4:2794 weakly excites
  14. L3:1472 weakly suppresses
  15. L4:1846 weakly excites
  16. L3:781 weakly excites
  17. L4:2367 weakly suppresses
  18. L4:738 weakly suppresses
  19. L1:2540 weakly excites
  20. L2:2396 weakly suppresses
  21. L4:2860 weakly suppresses
  22. L4:2652 weakly suppresses
  23. L3:810 weakly excites
  24. L2:2570 weakly suppresses

Feeds later layers

  1. L24:2031 weakly suppresses
  2. L7:1748 weakly excites
  3. L19:1233 weakly excites
  4. L9:1243 weakly suppresses
  5. L8:1798 weakly suppresses
  6. L8:795 weakly excites
  7. L11:1228 weakly excites
  8. L11:271 weakly excites
  9. L11:1103 weakly excites
  10. L12:2170 weakly suppresses
  11. L8:913 weakly suppresses
  12. L16:2872 weakly excites
  13. L11:604 weakly suppresses
  14. L16:1154 weakly excites
  15. L25:663 weakly excites
  16. L11:3024 weakly suppresses
  17. L10:2533 weakly excites
  18. L10:1539 weakly excites
  19. L8:1779 weakly excites
  20. L23:550 weakly suppresses
  21. L10:1447 weakly suppresses
  22. L21:99 weakly suppresses
  23. L8:3065 weakly excites
  24. L22:2200 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.

2 values between -1.19 and -1.093 values between -1.09 and -0.983 values between -0.98 and -0.882 values between -0.88 and -0.784 values between -0.78 and -0.683 values between -0.68 and -0.5710 values between -0.57 and -0.4713 values between -0.47 and -0.3754 values between -0.37 and -0.27253 values between -0.27 and -0.16785 values between -0.16 and -0.061673 values between -0.06 and 0.04783 values between 0.04 and 0.15208 values between 0.15 and 0.2560 values between 0.25 and 0.3519 values between 0.35 and 0.458 values between 0.45 and 0.5611 values between 0.56 and 0.661 values between 0.66 and 0.764 values between 0.76 and 0.862 values between 0.86 and 0.971 values between 1.07 and 1.173 values between 1.38 and 1.481 values between 1.68 and 1.791 values between 2.30 and 2.401 values between 2.81 and 2.92offnegativepositiveoften

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. -sc
  2. agination
  3. setDescription
  4. forward
  5. entieth
  6. olio
  7. survived
  8. -setup
  9. -over
  10. 毕竟是

Suppresses

  1. 使人
  2. .poi
  3. selon
  4. 怠
  5. derail
  6. stead
  7. reet
  8. nameof
  9. mü
  10. 使我