Neuron Explorer

Qwen3-0.6B

'same' in 'at the same time'

A neuron in layer 17 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 'same' in set phrases like 'at the same time' and 'in the same way'. When it was turned up, the model's text got a little more repetitive, but it did not show a clear theme tied to this phrase.

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

    Ars. "It started as an effort to prove we could create a commercial quality game using NASA content that is fun. NASA was looking for a project at the same time the Army Game Studio was looking to branch out from America’s Army. It

  2. #2 peak

    28-29 October 2015 in Luxembourg.⏎Today’s digital world offers children across Europe countless opportunities for personal growth and learning. At the same time, this digital world potentially exposes them to risks. We live in a time

  3. #3 peak

    throughout life. Brain trauma, vascular lesions, the effects of toxic substances such as alcohol, all of which accumulate over a lifetime and impact brain function. In the same way that some people see their hair just grow into adulthood while others keep a mop

  4. #4 peak

    950s and through scientific drilling in the oceans in the 1960s, the foundation for plate tectonics was laid at the same time, however, Wegener's groundbreaking theory was turned upside down.⏎Earthqu

  5. #5 peak

    pinpointing and eliminating habits that interfere with the productive use of time.⏎At the same time, many methods provide insight into the development of new and better habits, which

  6. #6 peak

    of the scavenge air amount, and thereby an increase of the exhaust gas temperature level, which leads to reduced efficiency of the diesel engine itself, at the same time creates a remarkable potential for increased power output on both power turbines and steam turbines

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: if the second, third, and, 4th, 5th, and, 5th, 6th, 6th, 6th, 7,7,7,6,6,6,6

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

    on these cloud fortresses, and will do so even more in years to come. Remarkable as the cloud is, there is no reason to think it is anywhere near completion. The generation of data and the need for its analysis continue to grow

  2. #2 peak

    ⏎The first gas masks were made in the late 1700s, and used exclusively by miners. These were very primitive, though. They were nowhere near as efficient as filtering out toxins as today's gas masks.⏎Gas masks were

  3. #3 peak

    er oscillator producing a sine wave of about 37 kHz, which is then boosted to about 2400 volts. That’s pretty good, but nowhere near the 15 kilovolts needed for a self

  4. #4 peak

    at http://www.fastmr.com/catalog/browse.aspx?category=16&page=1. The online catalog offers a description for each report, which can cost anywhere from $75.00 to $11,400.

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. L4:2695 suppresses
  2. L16:282 weakly excites
  3. L4:1505 weakly suppresses
  4. L7:1508 weakly excites
  5. L4:1330 weakly suppresses
  6. L16:80 weakly suppresses
  7. L6:2553 weakly suppresses 'for' after praise words like 'known for'
  8. L15:1145 weakly excites
  9. L3:323 weakly excites
  10. L3:1490 weakly suppresses
  11. L10:411 weakly suppresses
  12. L5:2924 weakly excites
  13. L5:888 weakly suppresses zero after a decimal point
  14. L1:1673 weakly suppresses
  15. L4:902 weakly excites
  16. L9:750 weakly excites
  17. L4:2453 weakly suppresses
  18. L16:1156 weakly excites
  19. L15:2846 weakly excites
  20. L16:2682 weakly suppresses
  21. L12:2855 weakly excites
  22. L11:1215 weakly suppresses
  23. L6:1351 weakly excites
  24. L7:1947 weakly excites

Feeds later layers

  1. L23:875 suppresses
  2. L26:551 suppresses
  3. L26:1060 weakly suppresses
  4. L18:1925 weakly excites
  5. L22:3027 weakly suppresses
  6. L19:2080 weakly excites ampersand joining two items
  7. L19:2201 weakly suppresses
  8. L19:933 weakly excites
  9. L18:2037 weakly suppresses
  10. L24:847 weakly suppresses
  11. L19:1183 weakly suppresses
  12. L19:643 weakly excites
  13. L23:742 weakly excites
  14. L23:1414 weakly suppresses
  15. L22:2966 weakly suppresses
  16. L21:2973 weakly suppresses
  17. L19:1070 weakly suppresses
  18. L25:2850 weakly suppresses
  19. L23:621 weakly suppresses
  20. L21:1242 weakly suppresses
  21. L18:726 weakly suppresses
  22. L18:1829 weakly excites
  23. L27:1896 weakly suppresses split names and comparative stems
  24. L25:2232 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.

4 values between -3.27 and -2.0041 values between -2.00 and -0.723758 values between -0.72 and 0.5673 values between 0.56 and 1.8413 values between 1.84 and 3.115 values between 3.11 and 4.396 values between 4.39 and 5.671 values between 6.94 and 8.222 values between 8.22 and 9.501 values between 10.78 and 12.051 values between 13.33 and 14.611 values between 22.27 and 23.551 values between 35.04 and 36.321 values between 46.54 and 47.81offnegativepositiveoften

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. principle
  2. underlying
  3. -sex
  4. ../../
  5. adesh
  6. iating
  7. هة
  8. ести
  9. 专业
  10. 原理

Suppresses

  1. rels
  2. asti
  3. differences
  4. atures
  5. Differences
  6. closer
  7. ACTER
  8. 更强
  9. cess
  10. esModule