Neuron Explorer

Qwen3-0.6B

"et" in "et al." citations

A neuron in layer 14 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 "et" in citations like "Smith et al.", the short form for "and others" after an author's name. When it was turned up, the model's output drifted into unrelated rambling about questions and problems, so the steering doesn't 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

    fever. The effect on mastoidectomy is unclear, but a clinically important increase seems unlikely. (Level of Evidence: 2c)⏎Sharland M, et al. Antibiotic prescribing in general practice and hospital admissions for peritonsill

  2. #2 peak

    into human 293T cells. The prominent spikes of the Env protein are visible under hight magnification. (Dewannieux, M. et al., Genome Res. (2006) 16: gr

  3. #3 peak

    froid and Delzenne (1998) suggest that inulin-type fructans may be⏎helpful for NIDD. Yamachita et al. (1984) observed a noticeable reduction⏎of cholesterolmia

  4. #4 peak

    influence of aerosols on the global climate.⏎Ault, A.P., et al. “Raman Microspectroscopy and Vibrational Sum Frequency Generation Spect

  5. #5 peak

    mathematics in the bilingual classroom, when the language of learning and teaching (LOLT) is not the learners’ mother tongue, is a complex issue (Setati et al 2002).⏎The purpose of this essay is to discuss the

  6. #6 peak

    ) and spruce (c. 70 years old) trees at the opposing sides of the plot (Figure 1a, for details see Grams et al. 2021). Six plots were equipped with roofs and assigned

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 want to test the hypothesis in the sample size (H2: α = α, α = α, α = α, α = α, α, α, α = α, α, α, α = α, α,

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

    for myself:⏎that for awhile I was the Heodeninga<|endoftext|>This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.⏎As the soil, however rich it may be, cannot

  2. #2 peak

    THE JEPSON MANUAL (1993)||⏎previous taxon |⏎Jepson Interchange (more information)⏎©Copyright 1993 by the Regents of the University of California⏎AND IS MAINTAINED FOR ARCHIVAL

  3. #3 peak

    Copyright © 1997, by Random House, Inc., on Infoplease.<|endoftext|>This was an introductory lesson

  4. #4 peak

    tionary, Creative Commons Attribution/Share-Alike License⏎- n. An underwater explorer.⏎from WordNet 3.0 Copyright 2006 by Princeton University. All rights reserved.⏎- n. an underwater swimmer equipped with

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. L13:1157 suppresses
  2. L10:546 excites
  3. L12:910 excites
  4. L12:967 excites
  5. L11:1327 suppresses
  6. L10:2446 excites
  7. L10:2112 excites
  8. L9:944 excites
  9. L13:1957 weakly suppresses
  10. L12:2887 weakly suppresses
  11. L7:211 weakly suppresses
  12. L6:2056 weakly excites
  13. L4:174 weakly suppresses
  14. L3:1918 weakly suppresses
  15. L5:1542 weakly excites
  16. L10:475 weakly excites
  17. L13:1978 weakly suppresses
  18. L2:831 weakly suppresses
  19. L7:2448 weakly suppresses
  20. L13:1635 weakly excites
  21. L12:1480 weakly suppresses
  22. L1:1392 weakly excites
  23. L10:2855 weakly excites
  24. L0:2711 weakly suppresses

Feeds later layers

  1. L15:1859 excites
  2. L16:940 suppresses
  3. L16:1322 excites
  4. L18:2235 excites
  5. L20:2181 excites
  6. L18:1306 weakly excites
  7. L18:56 weakly excites
  8. L15:636 weakly suppresses
  9. L16:1962 weakly excites
  10. L18:360 weakly excites
  11. L16:452 weakly excites
  12. L16:2867 weakly suppresses
  13. L18:953 weakly excites
  14. L16:320 weakly excites
  15. L16:996 weakly excites
  16. L21:617 weakly suppresses
  17. L16:1225 weakly suppresses
  18. L18:1486 weakly excites
  19. L15:432 weakly suppresses
  20. L17:295 weakly excites
  21. L15:870 weakly excites
  22. L19:2165 weakly excites
  23. L17:2126 weakly excites
  24. L19:237 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 -1.59 and -1.491 values between -1.29 and -1.197 values between -0.99 and -0.895 values between -0.89 and -0.7912 values between -0.79 and -0.6912 values between -0.69 and -0.5923 values between -0.59 and -0.4930 values between -0.49 and -0.3977 values between -0.39 and -0.29148 values between -0.29 and -0.19416 values between -0.19 and -0.091620 values between -0.09 and 0.01985 values between 0.01 and 0.11280 values between 0.11 and 0.21121 values between 0.21 and 0.3162 values between 0.31 and 0.4128 values between 0.41 and 0.5129 values between 0.51 and 0.6110 values between 0.61 and 0.7111 values between 0.71 and 0.817 values between 0.81 and 0.915 values between 0.91 and 1.014 values between 1.01 and 1.113 values between 1.11 and 1.212 values between 1.21 and 1.312 values between 1.31 and 1.411 values between 1.41 and 1.511 values between 1.61 and 1.712 values between 2.01 and 2.112 values between 2.11 and 2.211 values between 2.31 and 2.41offnegativepositiveoften

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. 彦
  2. ians
  3. 輯
  4. PTY
  5. icients
  6. .JSON
  7. 个多
  8. -blog
  9. тал
  10. арь

Suppresses

  1. tuyến
  2. intColor
  3. Divider
  4. füg
  5. 用工
  6. ToggleButton
  7. myModalLabel
  8. ửa
  9. 추
  10. uci