Neuron Explorer

Qwen3-0.6B

the word 'age' and 'how many'

A neuron in layer 5 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on the word "age" in phrases like "voting age", "age 13" or "age-related decline", and a little on "how many". The steering outputs don't show a clear link to it.

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

    graduating, she practiced in Germany where she also supported women doctors, helping to establish a women’s clinic and hospital.<|endoftext|>It's Hiking Season, But How Many Calories Are You Really Burning?⏎When the weather gets colder, one of the most

  2. #2 peak

    Dr. Nuha El Sayed: Thank you for having me.⏎So, today I would love to pick your brain about insulin resistance because it affects now how many tens of millions of people in our country, and then even more across the globe

  3. #3 peak

    aims to build the capacity and commitment of young Canadians to participate in their democracy. Student Vote's flagship program is a parallel election for students under the voting age, coinciding with official election periods. The program combines in-class learning, family

  4. #4 peak

    age 13 their minds and bodies are going through radical changes, which makes it difficult for them to focus on anything else. This might be too early an age to begin driving lessons, but

  5. #5 peak

    age, the authors report, while those who moved to the mainland — and consumed about 10 times less cocoa than their island-dwelling peers — showed age-related increases in blood

  6. #6 peak

    applied. A total of twenty nine depression patients in Sakura room were studied. Dependent t-test was applied to analize the data. The mayority of the age respondent between 16-24 years old. The male was fewer than

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: Like a person: age in number of years: 2023-07-10, 20:00:00-07:00, 2023:04:0

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

    regular picking: peas, beans, tomatoes, leafy crops, courgettes⏎3. WHEN? (see crop chart)⏎Right time: temperature, how long til harvest? Seed packets will tell you when to sow &

  2. #2 peak

    should be called getVolumeOfSphere(), take one parameter called radius and apply the relevant formula to calculate and return the volume of a sphere based on

  3. #3 peak

    efficiency is affected by angle of incline, characteristics of the specific bulk material, type of screw conveyor trough and screw pitch. Gkindus recommends designing screw conveyors using

  4. #4 peak

    tendency of different migrant groups to take out citizenship and to measure the size of groups eligible to vote. The data are useful cross-classified with birthplace, year of arrival in Australia and age data.⏎This page last updated 27 July

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. L2:324 suppresses 'less' in 'colorless'
  2. L4:379 excites
  3. L4:340 suppresses
  4. L4:1604 excites
  5. L4:299 excites
  6. L4:2497 suppresses
  7. L4:1065 excites
  8. L3:165 suppresses
  9. L4:3007 excites
  10. L0:2162 weakly suppresses
  11. L4:464 weakly suppresses
  12. L3:728 weakly excites
  13. L4:660 weakly excites
  14. L1:1529 weakly suppresses
  15. L4:2251 weakly excites
  16. L4:1385 weakly suppresses
  17. L3:354 weakly suppresses
  18. L3:2129 weakly suppresses
  19. L4:913 weakly excites
  20. L2:852 weakly suppresses
  21. L4:980 weakly excites
  22. L4:1217 weakly suppresses
  23. L3:1996 weakly suppresses
  24. L4:1791 weakly suppresses

Feeds later layers

  1. L25:1055 excites
  2. L8:1672 weakly suppresses
  3. L20:1891 weakly excites
  4. L7:1296 weakly excites
  5. L10:2887 weakly excites
  6. L7:423 weakly suppresses
  7. L8:392 weakly suppresses
  8. L8:382 weakly excites
  9. L6:686 weakly suppresses
  10. L12:2347 weakly suppresses
  11. L10:1197 weakly suppresses
  12. L14:3009 weakly suppresses
  13. L20:819 weakly excites
  14. L15:2244 weakly excites
  15. L17:1834 weakly suppresses
  16. L6:205 weakly excites
  17. L16:1091 weakly suppresses
  18. L8:2046 weakly excites
  19. L14:2302 weakly suppresses
  20. L22:1881 weakly suppresses
  21. L25:1645 weakly excites
  22. L13:1407 weakly excites
  23. L20:1844 weakly excites
  24. L18:1688 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.

2 values between -1.01 and -0.922 values between -0.84 and -0.752 values between -0.75 and -0.663 values between -0.66 and -0.582 values between -0.58 and -0.497 values between -0.49 and -0.4025 values between -0.40 and -0.3283 values between -0.32 and -0.23309 values between -0.23 and -0.15750 values between -0.15 and -0.061702 values between -0.06 and 0.03651 values between 0.03 and 0.11193 values between 0.11 and 0.2087 values between 0.20 and 0.2938 values between 0.29 and 0.3716 values between 0.37 and 0.468 values between 0.46 and 0.556 values between 0.55 and 0.638 values between 0.63 and 0.722 values between 0.72 and 0.813 values between 0.81 and 0.891 values between 0.89 and 0.982 values between 0.98 and 1.072 values between 1.15 and 1.241 values between 1.41 and 1.501 values between 1.67 and 1.761 values between 1.76 and 1.841 values between 2.36 and 2.45offnegativepositiveoften

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. !"; ⏎
  2. adero
  3. ugi
  4. ... ⏎
  5. elic
  6. ... ⏎
  7. 痘
  8. Magick
  9. edList
  10. íveis

Suppresses

  1. vant
  2. 网讯
  3. ivant
  4. ソン
  5. (insert
  6. це
  7. inch
  8. OkHttpClient
  9. ˣ
  10. lop