Neuron Explorer

Qwen3-0.6B

vague "certain" and "such" before a noun

A neuron in layer 8 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 lights up on words like "certain", "such", "various" and "specific" when they sit in front of a noun and mean some particular, unnamed one. Turning it up did not produce a clear change in topic or 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

    Every administrator and politician speaks of “party politics” in villages, and even villagers are often heard saying, “There is too much ‘party’ in such and such a village”. The coming of elections gives fresh opportunities for the crystallization of parties

  2. #2 peak

    14 in Senate primaries because of provisions in those states allowing for a second-round runoff election between the top two candidates if no candidate in the first round obtained a certain level of support.⏎Plurality Wins in Congressional Primaries, 19

  3. #3 peak

    there are perhaps more justifiable answers and less justifiable answers. Something like “Which of the following motivations best explains so-and-so’s actions in such-and-such novel?” Instead of posing this question, hoping that all your students take a moment

  4. #4 peak

    form ביון.⏎So what do לבייןand ביון mean, specifically? In mathematics there is a concept called interpolation, which involves estimating the location of various data points within a range – a precise, sophisticated game of connect the dots.

  5. #5 peak

    or bottom to find the answer. Students continue solving problems and finding answers as they move through the Fact Land. For some extra fun, students collect special tokens in certain boxes and tally them up at the bottom. Once students have made their way through

  6. #6 peak

    , subject to certain terms and conditions, the SEAI offers grants for both solar PV and EV charger

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 a 33-year-old, 20-year-old, 25-year-old, and 35-year-old. I am trying to get a group of 3 people together. So, how do I plan 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

    . In some cases, this step and hydrolysis, the next step, are combined into a single process.⏎In hydrolysis, the pretreated material, with the remaining solid carbohydrate fraction, primarily cellul

  2. #2 peak

    development of several species of seaweed using Lindenmayer systems (J.D. Corbit, D.J. Garbary). Generating fractals from Voronoi diagrams (K.W. Shirriff). Circles with kiss: a note

  3. #3 peak

    plane belongs to an algebraic curve if its coordinates satisfy a given polynomial equation. Basic questions involve the study of

  4. #4 peak

    ivia miniata: growing from seed can be difficult and it will be years before the plant will flower. An easier way to propagate Clivia miniata is by division, which can be done at any time of year. If you notice offsets (

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. L4:731 excites
  2. L7:832 weakly excites
  3. L7:52 weakly excites
  4. L1:1912 weakly excites
  5. L2:2512 weakly excites
  6. L6:186 weakly suppresses
  7. L3:1050 weakly excites
  8. L7:476 weakly suppresses
  9. L2:815 weakly suppresses
  10. L6:3045 weakly excites
  11. L7:1805 weakly excites
  12. L6:1813 weakly excites
  13. L4:1236 weakly suppresses
  14. L4:2251 weakly excites
  15. L7:2515 weakly suppresses
  16. L3:214 weakly suppresses
  17. L5:1972 weakly excites
  18. L5:283 weakly excites
  19. L5:172 weakly excites
  20. L5:32 weakly excites
  21. L6:1519 weakly suppresses
  22. L6:2654 weakly suppresses
  23. L4:892 weakly excites unclear: weak math-topic words
  24. L5:218 weakly excites

Feeds later layers

  1. L10:2231 suppresses
  2. L10:1966 weakly suppresses
  3. L17:1176 weakly excites
  4. L13:2026 weakly excites
  5. L12:777 weakly excites
  6. L21:2215 weakly suppresses
  7. L10:1031 weakly excites
  8. L16:1351 weakly excites
  9. L22:1728 weakly excites unclear: code assignments and parentheses
  10. L16:1128 weakly excites
  11. L10:455 weakly excites
  12. L23:2298 weakly suppresses
  13. L13:948 weakly suppresses
  14. L15:1640 weakly suppresses
  15. L24:1656 weakly suppresses
  16. L12:1750 weakly excites
  17. L13:1982 weakly excites
  18. L14:1373 weakly suppresses
  19. L14:880 weakly excites
  20. L11:1846 weakly suppresses
  21. L16:1139 weakly suppresses 'and' in lists of places or names
  22. L21:2634 weakly suppresses
  23. L24:2492 weakly suppresses
  24. L13:2081 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 -0.92 and -0.831 values between -0.83 and -0.741 values between -0.74 and -0.642 values between -0.64 and -0.554 values between -0.55 and -0.4628 values between -0.46 and -0.3795 values between -0.37 and -0.28245 values between -0.28 and -0.19549 values between -0.19 and -0.091100 values between -0.09 and -0.001113 values between -0.00 and 0.09452 values between 0.09 and 0.18204 values between 0.18 and 0.2775 values between 0.27 and 0.3621 values between 0.36 and 0.468 values between 0.46 and 0.554 values between 0.55 and 0.641 values between 0.82 and 0.912 values between 1.00 and 1.101 values between 1.37 and 1.461 values between 2.65 and 2.74offnegativepositiveoften

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. _JS
  2. outube
  3. QualifiedName
  4. ắp
  5. '../../../../
  6. _ABI
  7. بوك
  8. ../../
  9. finger
  10. pig

Suppresses

  1. rale
  2. elve
  3. seule
  4. TEE
  5. 不动
  6. Sixth
  7. otte
  8. 二线
  9. 客观
  10. otron