Neuron Explorer

Qwen3-0.6B

unclear: acronym and term starts in lists

A neuron in layer 24 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

It fires on the first letter or word of a technical term or acronym, often right after 'the', a bullet or a list marker, such as 'NIM' or 'NURBS'. Pushing it down made the model talk about costs, which this neuron seems to hold back.

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

    • Describe the characteristics of capital expenditure⏎• Describe the capital budgeting process⏎• Understand and apply the following capital budgeting techniques:⏎o Payback period⏎o Net present value⏎o Internal rate of return<|endoftext|>Becoming a Reader:

  2. #2 peak

    hallmark traits of PANS includes an abrupt onset of OCD or severely restricted food intake, in conjunction with the acute onset of several neuropsychiatric symptoms described in the NIM

  3. #3 peak

    function for the state n = 3, ℓ = 0, and mℓ = 0 is where σ = r/a0 and a0 is the Bohr radius (5.29 × 10-11 m

  4. #4 peak

    specifications for 304 Plate and 304L Stainless Steel Plate are ASTM A-240,ASME SA -240 and A666.⏎results for this questionWhat is better 3

  5. #5 peak

    architectural projects and even special effects in movies. It is essentially the evolution of a merger between a graphical user interface and Non-Uniform Rational B-Splines (NURBS) geometry, and currently functions for all major platforms.⏎Become

  6. #6 peak

    and Christmas cactus (Schlumbergera bridgesii)⏎Take a pinch of fertilizer and put in the bottle or

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: what if we *plan* *grees* *grees* *grees* *grees* *grees* *grees* *grees* *grees* *grees* *grees* *g

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

    entities to adjust their models accordingly and deliver optimized outputs and forecasts.⏎Digital twins are suitable for a wide range of applications, such as product lifecycle management (PLM), predictive maintenance of industrial machines, personnel training in virtual environments. In addition, digital

  2. #2 peak

    the total effect on Total Sales. However, if we want to know how the sales variance affects the profit, we use standard profit (absorption costing) or standard contribution marginal costing).⏎So, in order to calculate Sales Volume Variance, what

  3. #3 peak

    management framework means examining the entire range of business operations, identifying the strengths and weaknesses of current processes, and improving them. This includes individual workflows, administrative, support, and IT systems and the planning, production, and distribution of goods or services.

  4. #4 peak

    the standard price, standard profit or standard contribution.⏎This variance is quite important as it helps management assess how volume variance adversely affects the company’s performance. We use standard price if we want to assess how the variance affects total sales are.⏎We use

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. L20:340 suppresses
  2. L18:3032 suppresses
  3. L20:1035 excites
  4. L7:2972 excites
  5. L11:1489 excites
  6. L6:2007 excites
  7. L4:2237 weakly excites
  8. L9:507 weakly suppresses
  9. L4:2629 weakly suppresses
  10. L8:1548 weakly excites
  11. L21:873 weakly suppresses
  12. L7:2960 weakly excites
  13. L9:2356 weakly excites
  14. L3:1006 weakly excites
  15. L1:2577 weakly suppresses
  16. L22:159 weakly suppresses
  17. L21:2721 weakly suppresses
  18. L22:2056 weakly suppresses
  19. L3:1558 weakly suppresses
  20. L6:1145 weakly excites
  21. L20:443 weakly excites
  22. L10:1937 weakly excites
  23. L9:364 weakly suppresses
  24. L10:1283 weakly excites

Feeds later layers

  1. L27:2264 suppresses
  2. L27:1263 weakly suppresses
  3. L25:488 weakly suppresses
  4. L26:738 weakly excites
  5. L25:992 weakly suppresses
  6. L27:1911 weakly suppresses
  7. L25:906 weakly excites
  8. L27:2952 weakly suppresses
  9. L25:2343 weakly excites
  10. L26:1543 weakly suppresses
  11. L27:1195 weakly excites
  12. L25:124 weakly suppresses
  13. L26:2065 weakly suppresses
  14. L27:1919 weakly suppresses
  15. L25:1347 weakly suppresses the word 'sales'
  16. L27:871 weakly excites
  17. L25:230 weakly excites
  18. L25:1184 weakly suppresses
  19. L26:1482 weakly suppresses
  20. L25:439 weakly excites
  21. L27:1595 weakly suppresses
  22. L27:836 weakly suppresses
  23. L27:2064 weakly suppresses
  24. L27:1959 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 -22.50 and -21.711 values between -10.65 and -9.862 values between -7.49 and -6.702 values between -6.70 and -5.911 values between -5.91 and -5.124 values between -5.12 and -4.332 values between -4.33 and -3.549 values between -3.54 and -2.7514 values between -2.75 and -1.9639 values between -1.96 and -1.17536 values between -1.17 and -0.382819 values between -0.38 and 0.41409 values between 0.41 and 1.2027 values between 1.20 and 1.9916 values between 1.99 and 2.7810 values between 2.78 and 3.575 values between 3.57 and 4.364 values between 4.36 and 5.151 values between 5.15 and 5.934 values between 5.93 and 6.721 values between 6.72 and 7.511 values between 8.30 and 9.09offnegativepositiveoften

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. olds
  2. veyor
  3. ов
  4. chill
  5. vey
  6. �
  7. 冷水
  8. uler
  9. rá
  10. ظ

Suppresses

  1. 成本
  2. Cost
  3. cost
  4. 的成本
  5. cost
  6. costing
  7. Cost
  8. _cost
  9. Costs
  10. costs