Neuron Explorer

Qwen3-0.6B

the word before 'risk'

A neuron in layer 25 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 just before 'risk', as in 'at risk' or 'higher risk'. It appears to be predicting that the word 'risk' comes next. Turning it down made the model repeat 'risk' over and over, which shows how strongly it pushes that word.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. Think it is wrong? Flag it (opens an email to the maker with the neuron already filled in).

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

    how to care for people with disabilities, it promotes bias in healthcare. As a result, patients with disabilities lose trust in medical care, which can put their lives at risk.⏎The Impact of Low Accreditation Standards⏎On Dec. 14

  2. #2 peak

    a $195,000 Rapid Response Research grant from the National Science Foundation to, using machine learning, identify which COVID-19 patients are at risk of adverse cardiac events such as heart failure, sustained abnormal

  3. #3 peak

    , cardiogenic shock and death.⏎Increasing evidence of COVID-19’s negative impacts on the cardiovascular system highlights a great need for identifying COVID-19 patients at risk for heart problems, the researchers say. However, no such predictive capabilities currently

  4. #4 peak

    in the inner Bulgarian regions. Currently, an active surveillance system is in place for dairy factory employees and persons considered at risk after outbreaks in ruminants.⏎During 2005–2

  5. #5 peak

    desktop! more...⏎Orthodontics are often necessary to help improve the stability, function and health of an individual's teeth; otherwise, many people would be at higher risk for gum disease, tooth decay and tooth loss because of improper teeth positioning in

  6. #6 peak

    safely handle their vehicles and make prudent travel decisions. When bus drivers do not adhere to proper bus-handling protocol or traffic laws, they put dozens of vulnerable children at risk. The most common causes of fatal school-bus accidents include the following:⏎

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: risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk

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

    diligently to perform mitzvot (good deeds) to find favor with God during the 10 days of repentance between Rosh Hashanah and Yom Kippur.⏎According to Jewish teaching, God sits in judgment of the

  2. #2 peak

    to send the allied forces pell mell towards Berlin and get there before the Russians, Roosevelt was⏎adamant about sticking to the agreed upon plan conceived at the Yalta conference. Eisenhower would have liked ot send his forces to Berlin (Patton

  3. #3 peak

    J.M.; Mortimer, S.R.; Brereton, T.; Redhead, J.W

  4. #4 peak

    commemorate the Jewish holiday, Yom Ha’ Shoah, will begin at 5 p.m. in the

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. L17:2930 excites
  2. L22:2775 excites
  3. L19:2152 suppresses
  4. L23:2614 excites 'at' in 'at risk'
  5. L21:1439 suppresses
  6. L22:3060 excites
  7. L21:1887 suppresses
  8. L21:820 suppresses
  9. L18:2623 suppresses
  10. L15:2782 suppresses
  11. L21:1045 suppresses
  12. L20:1678 suppresses
  13. L21:1472 weakly suppresses
  14. L20:2973 weakly excites
  15. L23:1767 weakly excites
  16. L19:1461 weakly suppresses
  17. L21:2227 weakly excites
  18. L21:1699 weakly suppresses
  19. L19:2157 weakly suppresses
  20. L14:2691 weakly suppresses
  21. L22:1767 weakly suppresses
  22. L19:2900 weakly excites
  23. L17:3011 weakly suppresses
  24. L12:1715 weakly excites

Feeds later layers

  1. L26:1950 suppresses
  2. L27:221 weakly suppresses
  3. L26:2346 weakly suppresses
  4. L27:2273 weakly suppresses
  5. L27:2882 weakly suppresses
  6. L27:336 weakly suppresses
  7. L26:2591 weakly suppresses
  8. L26:2778 weakly suppresses
  9. L27:2396 weakly suppresses
  10. L27:2264 weakly suppresses
  11. L26:1495 weakly excites
  12. L26:1326 weakly suppresses
  13. L26:1314 weakly suppresses
  14. L27:2538 weakly suppresses
  15. L27:2695 weakly suppresses
  16. L26:2286 weakly suppresses
  17. L26:2205 weakly excites
  18. L26:2630 weakly suppresses
  19. L27:1759 weakly suppresses
  20. L26:2455 weakly suppresses
  21. L27:1630 weakly suppresses
  22. L26:1276 weakly suppresses
  23. L27:1191 weakly suppresses
  24. L26:627 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.

1 values between -8.89 and -7.016 values between -7.01 and -5.1212 values between -5.12 and -3.2447 values between -3.24 and -1.363190 values between -1.36 and 0.53619 values between 0.53 and 2.4112 values between 2.41 and 4.296 values between 4.29 and 6.182 values between 6.18 and 8.065 values between 8.06 and 9.943 values between 9.94 and 11.831 values between 13.71 and 15.592 values between 15.59 and 17.471 values between 26.89 and 28.771 values between 64.55 and 66.44offnegativepositiveoften

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. "));
  3. "));⏎
  4. ")); ⏎
  5. "));⏎⏎
  6. ")));⏎
  7. ")));
  8. defgroup
  9. ades
  10. ."));⏎

Suppresses

  1. risk
  2. 风险
  3. risk
  4. Risk
  5. Risk
  6. 的风险
  7. 風險
  8. risks
  9. -risk
  10. 风险管理