Neuron Explorer

Qwen3-0.6B

empty slot before a figure number

A neuron in layer 3 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 spot right after the word "Figure" or "Fig." in science writing, where the figure number is expected next. Turning it up did not produce any clear theme in the model's writing.

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

    cellular processes was cemented with the awarding of the 2004 Nobel Prize in Chemistry to ubiquitin research (3).Ubiquitin (Figure 1) can be attached to substrate proteins as a single moiety or in the

  2. #2 peak

    had proposed various methods of crowning that can be achieved in the process of gear generation. Maag engineers have used crowning for making longitudinal corrections (Fig. 1a); modifying involute tooth

  3. #3 peak

    is in place for dairy factory employees and persons considered at risk after outbreaks in ruminants.⏎During 2005–2007 (Figure 1), a total of 105 human cases of brucellos

  4. #4 peak

    RNAs were extracted from specimens using TRIzol reagent. Qiagen's miScript System was used to synthesize cDNA and perform qPCR (Figure 1). Endogenous miRNAs are not polyadenylated, therefore during

  5. #5 peak

    2 km. The infrasound data is collected at 20 hz, while the meteorologic data is recorded at 1 hz.⏎Figure 4. Array layout. Blue dots indicate the 7 elements, while the purple

  6. #6 peak

    ech (c. 90 years old) and spruce (c. 70 years old) trees at the opposing sides of the plot (Figure 1a, for details see Grams et al. 2021

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: it is not, but I have to be the one, and the one is the one. It is not, but I have to be the one, and the one is the one. It is not, but I have to be

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

    Resources to set limits that preserves the fishery. Indian fishing rights issues in the Northwest have been widely publicized over the years.⏎Alongside fishing rights, Amerindians also have the right to set up casinos and, much to my surprise,

  2. #2 peak

    0 to 1,000 years.⏎“What we can do is look at light bands in tree rings and correlate that to snowpack,” said Matheus. “Looking at snow, rather than runoff, is an important distinction because it

  3. #3 peak

    restricted by the installation of Road Speed Limiters. Road Speed Limiters were introduced in Europe in commercial trucks and passenger service vehicles initially to curb the maximum speed as this was seen as a major threat to road safety. While enforcing speed limits through fines

  4. #4 peak

    108], and in Arabic – in form ﺭﺎﻬ<|endoftext|>How Do I Score Well on the Spanish Praxis™?⏎The Spanish Praxis™ exam, more accurately known as the Praxis II®: Spanish: World

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. L0:2224 weakly suppresses
  2. L2:2328 weakly suppresses
  3. L2:513 weakly suppresses
  4. L2:1 weakly suppresses
  5. L0:786 weakly suppresses
  6. L0:1529 weakly excites
  7. L0:2189 weakly excites
  8. L2:1999 weakly excites
  9. L2:59 weakly excites
  10. L0:771 weakly excites
  11. L2:1927 weakly excites
  12. L0:3063 weakly excites
  13. L1:1373 weakly suppresses
  14. L0:1259 weakly excites
  15. L2:2500 weakly suppresses
  16. L1:904 weakly suppresses
  17. L0:2815 weakly suppresses
  18. L1:1779 weakly suppresses
  19. L0:2731 weakly excites
  20. L0:1102 weakly excites
  21. L1:323 weakly excites
  22. L2:36 weakly excites
  23. L2:2382 weakly excites
  24. L1:255 weakly excites "to" after "the best way is"

Feeds later layers

  1. L8:1186 suppresses
  2. L4:2839 suppresses
  3. L6:20 weakly suppresses
  4. L4:2950 weakly suppresses
  5. L17:1053 weakly suppresses
  6. L17:1438 weakly suppresses
  7. L9:450 weakly suppresses
  8. L21:2701 weakly suppresses
  9. L4:1802 weakly suppresses
  10. L11:176 weakly excites
  11. L6:1260 weakly suppresses
  12. L7:2638 weakly suppresses
  13. L13:665 weakly excites
  14. L23:2547 weakly suppresses
  15. L4:761 weakly suppresses
  16. L23:1634 weakly suppresses
  17. L24:1169 weakly suppresses
  18. L4:2043 weakly suppresses
  19. L16:761 weakly excites superlatives like 'largest' and 'first'
  20. L8:2853 weakly suppresses
  21. L4:604 weakly suppresses
  22. L8:1549 weakly suppresses
  23. L12:2056 weakly suppresses
  24. L15:150 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 -1.42 and -1.361 values between -1.36 and -1.291 values between -1.16 and -1.102 values between -1.04 and -0.971 values between -0.84 and -0.782 values between -0.78 and -0.712 values between -0.71 and -0.651 values between -0.65 and -0.583 values between -0.58 and -0.522 values between -0.52 and -0.459 values between -0.45 and -0.3920 values between -0.39 and -0.3340 values between -0.33 and -0.2698 values between -0.26 and -0.20237 values between -0.20 and -0.13576 values between -0.13 and -0.071052 values between -0.07 and -0.001104 values between -0.00 and 0.06410 values between 0.06 and 0.13210 values between 0.13 and 0.1974 values between 0.19 and 0.2624 values between 0.26 and 0.3211 values between 0.32 and 0.398 values between 0.39 and 0.458 values between 0.45 and 0.525 values between 0.52 and 0.581 values between 0.64 and 0.711 values between 0.71 and 0.772 values between 0.77 and 0.842 values between 1.10 and 1.16offnegativepositiveoften

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. ographed
  2. iculo
  3. icture
  4. .colorbar
  5. aghan
  6. ricula
  7. _sha
  8. quam
  9. ric
  10. illary

Suppresses

  1. scores
  2. 러
  3. NAS
  4. IEWS
  5. rat
  6. シーズン
  7. (HTTP
  8. ",⏎⏎
  9. '],['
  10. öst