Neuron Explorer

Qwen3-0.6B

'al' in 'et al.' citations

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 'al' in academic citations like 'Krebs et al., 1995'. It also reacts weakly to odd symbols in web addresses. Steering changed the topic a lot without any clear link to citations.

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

    regulation of prey numbers, population control of mesopredators through competition, and maintenance of a functional balance of biodiversity in the local community (Krebs et al., 1995; Terborgh et al., 19

  2. #2 peak

    08) is associated with marked ecological changes, including alterations to climatic regimes, phenology, resource availability and habitat quality (Zhou et al. 2004; Grimm et al. 2008

  3. #3 peak

    Floral humidity, an area of elevated humidity within the headspace of the flower, has been demonstrated to occur in a number of flower species (Corbet et al., 1979

  4. #4 peak

    European beech (Fagus sylvatica L.) are dominant tree species, accounting for 30% of the forest areas (Pretzsch et al. 2014). To elucidate the drought responses of these two

  5. #5 peak

    habitat, human-caused mortality, and isolation of small populations are major concerns in the conservation of large carnivores (Clark et al., 1996; Singleton & Lehmkuhl, 2

  6. #6 peak

    Hymenoptera: Vespidae) are widely distributed around most of the globe and are diverse and common in many landscapes (Beggs et al., 2011). Paper wasps are likely to influence many other

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 able to take the project in the present, but to do that, I have to know the future is coming, but in the future, the person is expected to have the knowledge to know the future, which is 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

    archive.jpr.org.uk/10.1177%2F13591045050537

  2. #2 peak

    and%20adaptive%20immune%20system.⏎- How Vitamin C Supports a Healthy Imm

  3. #3 peak

    > <H1>This is the Page Heading</H1> This is the first paragraph. <P> This is another paragraph

  4. #4 peak

    Python treats escape sequences in strings as special commands. For example, the `t` character inside a string prints a tab character (equivalent to four spaces):⏎s = “NametAgetMarks”⏎Name<|endoftext|>FAQ About Fashion⏎Cout

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. L18:1836 suppresses
  2. L15:1451 excites
  3. L9:722 suppresses
  4. L9:502 suppresses
  5. L4:2315 weakly suppresses
  6. L18:542 weakly suppresses
  7. L13:1411 weakly excites
  8. L12:2837 weakly excites
  9. L11:927 weakly excites
  10. L18:1128 weakly excites
  11. L0:148 weakly excites
  12. L12:95 weakly suppresses
  13. L15:2628 weakly excites
  14. L15:500 weakly excites
  15. L4:2208 weakly excites
  16. L11:1213 weakly excites
  17. L9:2228 weakly suppresses
  18. L8:2705 weakly excites
  19. L18:350 weakly excites
  20. L16:679 weakly suppresses
  21. L9:35 weakly excites
  22. L3:1278 weakly excites
  23. L2:573 weakly suppresses
  24. L2:2772 weakly suppresses

Feeds later layers

  1. L20:1092 excites
  2. L21:137 excites
  3. L20:573 excites
  4. L22:709 weakly excites
  5. L21:2775 weakly excites
  6. L20:2293 weakly excites
  7. L27:2426 weakly excites
  8. L21:322 weakly excites
  9. L27:452 weakly excites
  10. L22:1311 weakly suppresses
  11. L23:1945 weakly excites
  12. L22:856 weakly excites
  13. L23:2286 weakly excites
  14. L25:1398 weakly excites
  15. L27:2936 weakly excites
  16. L24:1850 weakly excites
  17. L26:2914 weakly excites
  18. L26:534 weakly excites
  19. L24:1697 weakly excites
  20. L22:2075 weakly excites
  21. L20:3034 weakly excites
  22. L20:2160 weakly excites
  23. L22:2383 weakly excites
  24. L20:2093 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.

2 values between -2.41 and -1.847 values between -1.84 and -1.26217 values between -1.26 and -0.681860 values between -0.68 and -0.111692 values between -0.11 and 0.4796 values between 0.47 and 1.0414 values between 1.04 and 1.629 values between 1.62 and 2.202 values between 2.20 and 2.774 values between 2.77 and 3.351 values between 3.92 and 4.501 values between 4.50 and 5.081 values between 5.08 and 5.651 values between 12.56 and 13.141 values between 20.05 and 20.63offnegativepositiveoften

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. rowser
  2. IZED
  3. ized
  4. icrobial
  5. iddle
  6. €™
  7. ISED
  8. IDDLE
  9. ruled
  10. ум

Suppresses

  1. DataContext
  2. /bind
  3. ystate
  4. Paw
  5. Dane
  6. trys
  7. 实践
  8. técnica
  9. 表白
  10. _va