Neuron Explorer

Qwen3-0.6B

"of" after is/are/be (of + quality)

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 lights up on the word "of" when it comes right after a linking verb like "is", "are" or "be", as in "are of considerable importance" or "of two types". When it was turned up, the model's writing did not become clearly about anything in particular, so the steering results do not show a clear effect.

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

    ) by rotating a magnetic field at a constant angular velocity within a stationary coil as shown in Fig. (a).⏎In either case, the generated voltage will be of sinusoidal waveform. The magnitude of generated voltage will

  2. #2 peak

    axes is perpendicular⏎to other axes and is of different length, e.g.⏎(iv) Orthorhombic System :⏎Three

  3. #3 peak

    is available for this purpose (e.g., US Census data at https://www.census.gov/) and can be of tremendous help in targeting various ethnic groups and geographic areas with specific fruit crops. Additionally

  4. #4 peak

    blackens the leaves; pull up the plants and discard them.<|endoftext|>Introduction and general survey⏎The economic importance of power amplifiers⏎Audio power amplifiers are of considerable economic importance. They are built in their hundreds of thousands every year, and

  5. #5 peak

    and should be of sufficient length to penetrate the muscle tissue. The anterolateral thigh can<|endoftext|>

  6. #6 peak

    interests of children in various age groups in order to write games that really appeal to the audience. In children's literature, for example, the protagonists tend to be of a similar age to their intended audience, a theme that game studios can apply to

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 2011 Mac 6000 (the first time I have a 6000) with a 60000-based 6000 (the first time I have 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

    blood sugar. This can put you more at risk for Type 2 diabetes. This is one more reason that avoiding food all day and eating all of your calories near bedtime is not a good idea.⏎Another pattern that exists for people who eat late

  2. #2 peak

    and run without help. Enjoy every minute.⏎- FACT: 95 percent of 6- to 21-year-old students with disabilities attended public schools in 2007.⏎- TIP: If the baby does not crawl

  3. #3 peak

    TMJD as well.⏎What Causes Pain in the Temporomandibular Joints?⏎The temporomandibular joints (TMJs) are located in each ear, between the jawbone and skull. When you chew, your mouth

  4. #4 peak

    oceans. With over 80% of ocean debris ending in the ocean water, it impacts marine organisms. If not taken care of, plastics

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. L16:2035 excites 'of' after a noun phrase
  2. L15:2682 excites
  3. L16:2291 suppresses
  4. L15:1223 weakly suppresses
  5. L6:2310 weakly suppresses
  6. L7:2705 weakly suppresses
  7. L12:2106 weakly suppresses
  8. L14:1492 weakly excites
  9. L15:1683 weakly suppresses
  10. L11:2840 weakly excites 'in' after a size measurement
  11. L15:2180 weakly suppresses
  12. L10:132 weakly excites
  13. L6:2156 weakly suppresses
  14. L6:1343 weakly suppresses
  15. L5:2462 weakly excites
  16. L5:47 weakly excites
  17. L16:2307 weakly excites
  18. L16:2564 weakly suppresses
  19. L16:1557 weakly excites 'in' meaning 'in this case' or 'in between'
  20. L5:1116 weakly suppresses
  21. L2:1323 weakly suppresses unclear: fragments of rare nouns
  22. L0:582 weakly suppresses
  23. L4:862 weakly suppresses 'to' after 'too big' or 'right angles'
  24. L2:851 weakly suppresses

Feeds later layers

  1. L18:1468 excites
  2. L18:1770 excites
  3. L19:1610 excites
  4. L22:1073 weakly excites
  5. L20:1494 weakly excites
  6. L21:444 weakly excites
  7. L19:964 weakly excites
  8. L19:248 weakly suppresses
  9. L19:769 weakly excites
  10. L18:1007 weakly excites
  11. L25:339 weakly excites
  12. L19:178 weakly excites
  13. L24:2680 weakly suppresses
  14. L18:1046 weakly excites
  15. L19:836 weakly excites
  16. L18:2885 weakly suppresses
  17. L19:1843 weakly suppresses
  18. L18:1690 weakly excites
  19. L21:662 weakly excites
  20. L19:2064 weakly excites
  21. L22:1689 weakly excites
  22. L18:1271 weakly excites
  23. L21:171 weakly excites
  24. L20:161 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 -6.12 and -5.571 values between -5.03 and -4.483 values between -4.48 and -3.943 values between -3.94 and -3.406 values between -3.40 and -2.8511 values between -2.85 and -2.319 values between -2.31 and -1.7625 values between -1.76 and -1.2266 values between -1.22 and -0.67741 values between -0.67 and -0.132702 values between -0.13 and 0.42289 values between 0.42 and 0.9619 values between 0.96 and 1.5112 values between 1.51 and 2.056 values between 2.05 and 2.602 values between 3.14 and 3.682 values between 3.68 and 4.232 values between 5.32 and 5.862 values between 6.41 and 6.952 values between 8.59 and 9.131 values between 9.67 and 10.221 values between 12.94 and 13.491 values between 14.58 and 15.121 values between 15.12 and 15.66offnegativepositiveoften

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. entic
  2. verty
  3. чик
  4. awai
  5. atever
  6. alf
  7. enty
  8. posal
  9. inox
  10. etry

Suppresses

  1. 妞
  2. Die
  3. well
  4. bien
  5. アウト
  6. Phantom
  7. ('@
  8. 出局
  9. 敢
  10. 胼