Neuron Explorer

Qwen3-0.6B

sentence-opening 'indeed' and similar

A neuron in layer 16 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on connecting words that start a new point or add support, such as 'Indeed', 'Not only' and sometimes the end of a sentence just before one. The pattern is not very tight, and turning the neuron up did not produce a clear theme.

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

    they are not static. They can change profoundly throughout your life, for better or for worse, and one of the biggest influences on this change is your diet.⏎Indeed

  2. #2 peak

    , carbon dioxide, sulfur, and microorganisms. (Crude oil with lots of sulfur in it is called “sour,” as opposed to “sweet.”) If enough of these contaminants collect in a steel pipe, they’ll work together to

  3. #3 peak

    acy often decided to sacrifice the fate of other nations within the Republic.⏎This phenomenon occurred with varying intensity and, sadly, it is apparently being revived today. Not only in the form of Pope Francis’ unfortunate statements but also in the form of another

  4. #4 peak

    exercise, and plain-old-fashioned muscle fatigue. In addition, a lack of flexibility in the hamstrings is thought to spike the risk of hamstring injury (2). Indeed, research has suggested that athletes who have a history of hamstring injury have significantly less

  5. #5 peak

    tasks allowing your workforce to focus on more demanding tasks.⏎Also, these robots are very versatile and can quickly be relocated to another part of your production process. ndeed, compared to larger complex solutions such as traditional robots, cobots are quick to

  6. #6 peak

    beef requires 40 calories of fossil fuel energy, whilst producing one calorie of human-edible grain takes only 2.2 calories of fossil fuel energy4!⏎2) Preserve Habitats and Species⏎Eating animals is

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: the fact that, in reality, the fact that the number of times that we can take the number of times we can take that number of times that number of times that number of times that number of times that number of times that number of

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

    added as we transcribe it.⏎- Front Matter⏎- A PREFACE containing a general Account of the Life, Character, and Writings of the AUTHOR: by John Clarke⏎- I. Of Moral Good and Evil⏎- II.

  2. #2 peak

    its disorders. In: Daroff RB, Fenichel GM, Jankovic J, Mazziotta JC, eds. Bradley's Neurology in Clinical Practice. 6th ed. Philadelphia, PA: Elsevier Saunders; 20

  3. #3 peak

    the the ptr array are interpreted as unsigned char for the comparison.⏎- Pointer to the block of memory where the search is performed.⏎- Value to be located. The value is passed

  4. #4 peak

    usion. In: Mason RJ, Murray JF, Broaddus VC, et al., eds. Murray and Nadel's Textbook of Respiratory Medicine. 5th ed.

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. L14:1209 excites
  2. L9:12 suppresses
  3. L9:16 suppresses unclear: periods inside numbers and names
  4. L15:8 excites
  5. L13:212 weakly suppresses
  6. L14:602 weakly suppresses
  7. L10:144 weakly suppresses
  8. L12:731 weakly excites
  9. L11:112 weakly excites
  10. L15:1283 weakly suppresses
  11. L14:1187 weakly suppresses
  12. L4:96 weakly excites
  13. L14:451 weakly excites
  14. L5:1653 weakly excites
  15. L5:3 weakly suppresses
  16. L5:974 weakly excites
  17. L2:1321 weakly excites
  18. L6:1304 weakly suppresses
  19. L5:1371 weakly suppresses
  20. L12:653 weakly excites
  21. L13:9 weakly excites
  22. L8:1683 weakly suppresses
  23. L13:345 weakly excites
  24. L7:948 weakly suppresses

Feeds later layers

  1. L17:1394 weakly suppresses
  2. L22:3018 weakly excites
  3. L17:806 weakly excites
  4. L19:1638 weakly suppresses
  5. L18:959 weakly suppresses
  6. L20:2711 weakly suppresses
  7. L20:241 weakly suppresses
  8. L17:2852 weakly suppresses
  9. L22:2981 weakly suppresses
  10. L20:3021 weakly suppresses
  11. L19:2928 weakly suppresses
  12. L18:2242 weakly suppresses
  13. L23:2679 weakly excites
  14. L21:1594 weakly suppresses
  15. L21:578 weakly excites
  16. L25:1405 weakly excites
  17. L17:503 weakly suppresses
  18. L17:2285 weakly excites
  19. L19:2762 weakly suppresses
  20. L23:2362 weakly suppresses
  21. L21:724 weakly suppresses
  22. L18:2954 weakly suppresses
  23. L17:2499 weakly suppresses
  24. L17:1726 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.

1 values between -4.43 and -4.161 values between -3.35 and -3.081 values between -3.08 and -2.811 values between -2.81 and -2.543 values between -2.54 and -2.278 values between -2.27 and -2.004 values between -2.00 and -1.7316 values between -1.73 and -1.4617 values between -1.46 and -1.2039 values between -1.20 and -0.9364 values between -0.93 and -0.66163 values between -0.66 and -0.39803 values between -0.39 and -0.121794 values between -0.12 and 0.16391 values between 0.16 and 0.43151 values between 0.43 and 0.7096 values between 0.70 and 0.9773 values between 0.97 and 1.2460 values between 1.24 and 1.5151 values between 1.51 and 1.7840 values between 1.78 and 2.0528 values between 2.05 and 2.3124 values between 2.31 and 2.5820 values between 2.58 and 2.8518 values between 2.85 and 3.1214 values between 3.12 and 3.396 values between 3.39 and 3.663 values between 3.66 and 3.934 values between 3.93 and 4.204 values between 4.20 and 4.471 values between 4.47 and 4.744 values between 4.74 and 5.012 values between 5.01 and 5.281 values between 5.28 and 5.551 values between 5.55 and 5.821 values between 6.09 and 6.36offnegativepositiveoften

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. (del
  2. rdf
  3. ński
  4. _PARAMETER
  5. rien
  6. enze
  7. 〘
  8. encing
  9. Raises
  10. plier

Suppresses

  1. correctly
  2. thăm
  3. 경우에는
  4. 括
  5. yyn
  6. uya
  7. 情况下
  8. 際に
  9. erva
  10. Brook