Neuron Explorer

Qwen3-0.6B

'organisms' and 'Code' in reference contexts

A neuron in layer 6 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron responds to a handful of words that name things in science and law, such as 'organisms', 'Code' in 'United States Code', and 'object'. Turning it up did not produce a clear topic in the writing.

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

    § 1395 tells us that this cited federal statute can be found in Title 42, Section 1395 of the United States Code.⏎Code: In general, the term "code" refers to the main body

  2. #2 peak

    to the same ___________.⏎In the scientific name of the white oak, Quercus alba, Quercus is the __________________.⏎Two groups of organisms that are farther from each other on a cladogram share __________ derived traits than

  3. #3 peak

    Revised Statutes). The statutes that are published in a state’s code are grouped by subject matter into Titles, as in Title 11 of the United States Code (bankruptcy laws). In some states, including California, Texas and New York

  4. #4 peak

    buses.⏎- Cars, trucks and motorcycles produce three times the pollution of buses.⏎These comparisons assume the vehicles are not filled to capacity on average. Trains come out even better in comparison to planes when both are full. (Note that Amtrak

  5. #5 peak

    history.⏎How did Aristole classify life? What was the problem with his system?⏎Scientist should use scientific names because common names can be misleading due to organims having multiple names. One type of species can live all over the world and can

  6. #6 peak

    An element can be represented on several diagrams. You can quickly locate which diagrams it occurs in throughout the model,

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: a part of a part of the system. I (I) can't't. And A.2553355 256, 700, 700, 700

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

    - 4 Clinical Trials⏎- 5 Cohort (Longitudinal) Studies⏎- 6 Case-Control Studies⏎- 7 Mesures of Morbidity and Mortality⏎Fundamentals in Epidemiology Course Description⏎Fundamentals of

  2. #2 peak

    %, respectively, of all cases reported by EU/EEA countries. 31 deaths were reported for this period.⏎Of 13,783 cases with known vaccination status, 87% were unvaccinated,

  3. #3 peak

    1971 and 9.8 million in 2001, according to a report published in the March 11 issue of Morbidity and Mortality Weekly Report.⏎And the number of cancer survivors continues to grow because

  4. #4 peak

    journey in a computer graphics world (C.A. Pickover). Part II. Attractors. Automatic generation of strange attractors (J.C. Sprott). Attractors with dueling symmetry (C.A. Reiter). A

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. L5:216 suppresses
  2. L4:200 excites
  3. L5:264 suppresses
  4. L5:1939 suppresses
  5. L5:2709 suppresses
  6. L4:2695 weakly excites
  7. L4:867 weakly excites
  8. L4:190 weakly suppresses
  9. L5:2190 weakly suppresses
  10. L5:314 weakly suppresses
  11. L3:393 weakly excites
  12. L5:2937 weakly suppresses
  13. L4:215 weakly excites
  14. L4:151 weakly suppresses
  15. L4:517 weakly excites
  16. L4:2103 weakly suppresses
  17. L3:390 weakly excites
  18. L5:28 weakly excites
  19. L4:272 weakly excites
  20. L1:2699 weakly suppresses
  21. L5:60 weakly excites
  22. L4:2459 weakly suppresses
  23. L4:37 weakly suppresses
  24. L2:480 weakly suppresses

Feeds later layers

  1. L10:1073 suppresses
  2. L15:2846 suppresses
  3. L8:120 excites
  4. L8:1311 excites
  5. L7:769 suppresses
  6. L15:1044 weakly suppresses
  7. L14:1760 weakly suppresses
  8. L10:1507 weakly suppresses
  9. L11:2402 weakly suppresses
  10. L9:620 weakly suppresses
  11. L8:2240 weakly suppresses
  12. L12:2098 weakly suppresses
  13. L9:707 weakly suppresses
  14. L13:705 weakly excites
  15. L7:116 weakly suppresses
  16. L9:1289 weakly suppresses
  17. L7:62 weakly suppresses
  18. L17:2111 weakly suppresses
  19. L14:1587 weakly suppresses
  20. L8:845 weakly suppresses 'too' meaning 'also' at clause end
  21. L11:2685 weakly suppresses
  22. L16:3042 weakly suppresses
  23. L14:2846 weakly suppresses
  24. L12:608 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.

4 values between -0.71 and -0.662 values between -0.66 and -0.625 values between -0.62 and -0.5718 values between -0.57 and -0.5320 values between -0.53 and -0.4835 values between -0.48 and -0.4446 values between -0.44 and -0.3988 values between -0.39 and -0.34112 values between -0.34 and -0.30163 values between -0.30 and -0.25245 values between -0.25 and -0.21316 values between -0.21 and -0.16409 values between -0.16 and -0.12472 values between -0.12 and -0.07410 values between -0.07 and -0.02491 values between -0.02 and 0.02300 values between 0.02 and 0.07207 values between 0.07 and 0.11150 values between 0.11 and 0.16102 values between 0.16 and 0.2176 values between 0.21 and 0.2555 values between 0.25 and 0.3040 values between 0.30 and 0.3431 values between 0.34 and 0.3917 values between 0.39 and 0.439 values between 0.43 and 0.4815 values between 0.48 and 0.5313 values between 0.53 and 0.5713 values between 0.57 and 0.625 values between 0.62 and 0.6610 values between 0.66 and 0.717 values between 0.71 and 0.756 values between 0.75 and 0.804 values between 0.80 and 0.852 values between 0.85 and 0.891 values between 0.89 and 0.944 values between 0.94 and 0.981 values between 0.98 and 1.033 values between 1.03 and 1.071 values between 1.07 and 1.12offnegativepositiveoften

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. cbd
  2. cb
  3. cbc
  4. renders
  5. �
  6. sh
  7. sel
  8. mounts
  9. md
  10. .RequestBody

Suppresses

  1. phia
  2. async
  3. Exited
  4. Caught
  5. extrême
  6. Moder
  7. 都要
  8. 厚厚的
  9. esters
  10. gone