Neuron Explorer

Qwen3-0.6B

'the' in mid-sentence noun phrases

A neuron in layer 19 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 'the' when it appears in the middle of a sentence and a noun or adjective is expected next. The steering results do not show a clear theme, so it is hard to say what the neuron does beyond that.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

    , Wegener's groundbreaking theory was turned upside down.⏎Earthquakes are not only terrible natural disasters, they also offer a view inside the Earth. It was the geophysicists Wadati and Benioff, who in 1

  2. #2 peak

    and you’re all set.⏎At one time, all engine blocks were cast iron. Today, they’re aluminum. Be it iron or aluminum, the process is the same, minus the melting point.⏎What is Sintered Metal?⏎Before any

  3. #3 peak

    provide the same amount of heat to the product as the boiling water processes. Just because pressure is used to decrease the process time, the canning process is not the same as one to destroy spores of Clostridium botulinum as

  4. #4 peak

    or rag. These low-tech lamps smoked and gave off faint, flickering light, similar to candles. Candles and devices using this oil-burning technique were the main means used to light homes, employed in candelabras, wall scon

  5. #5 peak

    OCT uses light waves for imaging. The shorter wavelength of light empowers this technology to generate high-resolution, cross-sectional images with zero delays.⏎Time Domain technology was the first method that was used in OCT technology. Time Domain technology uses a low-co

  6. #6 peak

    iniae cause severe intestinal infections. Studies using Yersinia pseudotuberculosis as a model organism aim to elucidate the infection mechanisms of these...⏎Researchers from the

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 to be very careful, and I have to be very careful, and I have to be very careful, I have to, I have to, and I have to. (I have been writing this, but I have written,

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

    Guide⏎The Production Process⏎- Domestic Resource and Production Projections⏎- Outlook for Natural Gas Prices⏎- Economic Benefits and Implications⏎- Public Health and the Environment⏎States Take Action: The Balancing Act⏎State Policy Actions⏎-

  2. #2 peak

    -interaction. The students are also encouraged to seek out and take advantage of other opportunities that will further their expertise: computer-related part-time employment, internships, the Computer Science club, field trips, invited speakers, LAN parties :), etc.

  3. #3 peak

    JSTOR JSTOR is part of ITHAKA, a not-for-profit organization helping the academic community use digital technologies to preserve the scholarly record and to advance research and teaching in

  4. #4 peak

    justice policies.. All of these plans are informed by the Federal GND — they are explicit plans to address not only the climate crisis but also racial justice, the housing crisis, the pandemic and a Just<|endoftext|>Wheels, like other load-bearing components

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. L18:171 strongly excites
  2. L15:888 strongly excites
  3. L18:1359 strongly excites
  4. L16:652 strongly suppresses
  5. L11:1168 strongly excites
  6. L9:662 strongly suppresses
  7. L18:641 suppresses
  8. L16:1594 excites
  9. L10:120 suppresses
  10. L18:1913 suppresses
  11. L18:1225 suppresses
  12. L18:216 excites
  13. L15:1888 excites
  14. L18:2630 suppresses
  15. L10:1377 suppresses
  16. L17:30 suppresses
  17. L10:321 suppresses
  18. L10:1644 excites
  19. L9:227 excites
  20. L4:73 suppresses
  21. L3:344 excites
  22. L1:740 suppresses
  23. L9:2198 excites
  24. L16:792 excites "the" inside a prepositional phrase

Feeds later layers

  1. L25:377 excites
  2. L24:109 suppresses
  3. L20:1047 excites
  4. L21:675 excites
  5. L22:260 suppresses
  6. L24:153 excites
  7. L20:1989 suppresses
  8. L20:240 suppresses
  9. L23:218 suppresses
  10. L27:60 suppresses
  11. L22:1081 suppresses
  12. L25:335 excites
  13. L26:111 excites
  14. L22:211 suppresses
  15. L25:390 suppresses
  16. L20:2117 excites
  17. L24:360 suppresses
  18. L23:79 excites
  19. L24:434 excites
  20. L25:812 excites
  21. L20:889 excites
  22. L26:552 excites
  23. L20:2092 excites
  24. L20:932 suppresses 'of' after a count or quantity

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 -10.34 and -9.981 values between -9.98 and -9.621 values between -8.90 and -8.541 values between -8.54 and -8.171 values between -8.17 and -7.812 values between -7.81 and -7.451 values between -7.45 and -7.091 values between -6.73 and -6.374 values between -6.37 and -6.002 values between -6.00 and -5.641 values between -5.64 and -5.282 values between -5.28 and -4.928 values between -4.92 and -4.563 values between -4.56 and -4.203 values between -4.20 and -3.835 values between -3.83 and -3.477 values between -3.47 and -3.116 values between -3.11 and -2.7514 values between -2.75 and -2.3919 values between -2.39 and -2.0219 values between -2.02 and -1.6621 values between -1.66 and -1.3037 values between -1.30 and -0.9443 values between -0.94 and -0.58238 values between -0.58 and -0.222110 values between -0.22 and 0.151026 values between 0.15 and 0.51231 values between 0.51 and 0.8731 values between 0.87 and 1.2318 values between 1.23 and 1.5911 values between 1.59 and 1.967 values between 1.96 and 2.3215 values between 2.32 and 2.687 values between 2.68 and 3.045 values between 3.04 and 3.403 values between 3.40 and 3.763 values between 3.76 and 4.13offnegativepositiveoften

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. decisions
  2. predictions
  3. transactions
  4. reactions
  5. instances
  6. discussions
  7. responses
  8. examples
  9. seminars
  10. figures

Suppresses

  1. epad
  2. 印象深刻
  3. ��
  4. exporter
  5. ductor
  6. odus
  7. rometer
  8. atum
  9. ocese
  10. classpath