Neuron Explorer

Qwen3-0.6B

math word problem wording

A neuron in layer 15 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

It fires weakly on the wording of maths and word problems, such as 'in a geometric sequence', 'the same' or the full stop after a number in a practice question. It is quiet overall, and turning it up produced vague, story-like text rather than maths.

Written by Claude from this neuron's activations and steering outputs. Confidence: medium. 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

    make at the end of the third week?⏎3. Solve for the following.⏎a. In a geometric sequence, and . Find .⏎b. In a geometric sequence, and . Find .⏎In

  2. #2 peak

    One of the most common uses of this calculation is when purchasing a product on sale.⏎Thus, if the marked price is 1,000 and has a discount rate of 25% should take the following relationship: 1.

  3. #3 peak

    model within an Earth system model, we analysed the yield responses of six major crops to three SG technologies (SGs) and emissions reduction when they provide roughly the same reduction in radiative forcing

  4. #4 peak

    same slope, m = 2, and we know that,⏎Lines with the same slope are parallel.⏎Hence, the given lines

  5. #5 peak

    elementary problem solving series into printable worksheets:⏎Milk Street — Boston, 1706⏎Ben helped his father make 650 tallow candles. After selling some, they had 39 candles left.

  6. #6 peak

    . Supported by the Birmingham Buddhist Vihara, the pagoda underwent a large amount of planning. The circumference of the pagoda is 153 ft. and is divided into eight sections. The pagoda is in the Burmese

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: if I were a question to the person that I would be a question to the person that I would be a question to the person that I would be a question to the person that I would be a question to the person that I would be

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

    of 0.02). Plug your population size and required margin of error into the formula. The result will be the number of samples you need to take.⏎For example, suppose that you have a

  2. #2 peak

    make at the end of the third week?⏎3. Solve for the following.⏎a. In a geometric sequence, and . Find .⏎b.

  3. #3 peak

    liu.<|endoftext|>Bond Price Calculator⏎With this bond price calculator, we aim to help you calculate the bond price issued by a government or a corporation. Finding out the current bond price

  4. #4 peak

    = 9 – 5.⏎Made = 650⏎Sold = ?⏎Leftover = 39⏎Leftover = Made – Sold⏎Which means also that:⏎Sold = Made – Leftover⏎Sold = 650

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. L13:1197 excites
  2. L14:265 suppresses
  3. L14:1982 suppresses
  4. L14:1417 excites
  5. L13:2962 excites
  6. L12:2199 excites
  7. L14:2911 excites
  8. L12:2945 suppresses
  9. L11:346 weakly excites
  10. L12:528 weakly excites
  11. L14:1968 weakly suppresses
  12. L12:2026 weakly suppresses
  13. L11:143 weakly excites
  14. L9:1978 weakly suppresses
  15. L3:465 weakly excites
  16. L12:471 weakly suppresses
  17. L11:1599 weakly suppresses
  18. L12:268 weakly excites
  19. L14:368 weakly suppresses
  20. L11:1992 weakly excites
  21. L9:91 weakly excites
  22. L14:1053 weakly suppresses
  23. L8:808 weakly excites
  24. L13:2577 weakly excites

Feeds later layers

  1. L16:2999 suppresses
  2. L16:2846 suppresses
  3. L17:1220 suppresses
  4. L19:3020 suppresses
  5. L21:2920 excites
  6. L16:887 weakly suppresses
  7. L16:982 weakly suppresses
  8. L16:2633 weakly suppresses
  9. L18:763 weakly suppresses
  10. L19:1488 weakly suppresses
  11. L16:1593 weakly suppresses
  12. L17:2815 weakly suppresses
  13. L20:1996 weakly suppresses
  14. L17:1449 weakly excites
  15. L20:2260 weakly excites
  16. L16:2175 weakly suppresses
  17. L18:2259 weakly suppresses
  18. L19:2174 weakly suppresses
  19. L18:2672 weakly suppresses
  20. L19:2844 weakly suppresses
  21. L21:2227 weakly suppresses
  22. L17:1019 weakly suppresses
  23. L24:2496 weakly suppresses
  24. L16:2961 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 -0.61 and -0.573 values between -0.57 and -0.544 values between -0.54 and -0.515 values between -0.51 and -0.4811 values between -0.48 and -0.4512 values between -0.45 and -0.4124 values between -0.41 and -0.3844 values between -0.38 and -0.3559 values between -0.35 and -0.3288 values between -0.32 and -0.28101 values between -0.28 and -0.25128 values between -0.25 and -0.22186 values between -0.22 and -0.19207 values between -0.19 and -0.15257 values between -0.15 and -0.12277 values between -0.12 and -0.09297 values between -0.09 and -0.06323 values between -0.06 and -0.03347 values between -0.03 and 0.01318 values between 0.01 and 0.04266 values between 0.04 and 0.07213 values between 0.07 and 0.11184 values between 0.11 and 0.14151 values between 0.14 and 0.17123 values between 0.17 and 0.2086 values between 0.20 and 0.2371 values between 0.23 and 0.2743 values between 0.27 and 0.3029 values between 0.30 and 0.3318 values between 0.33 and 0.367 values between 0.36 and 0.409 values between 0.40 and 0.436 values between 0.43 and 0.464 values between 0.46 and 0.491 values between 0.49 and 0.532 values between 0.53 and 0.561 values between 0.56 and 0.591 values between 0.65 and 0.69offnegativepositiveoften

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. .Hit
  2. raphics
  3. -render
  4. utive
  5. ERRUPT
  6. fried
  7. enie
  8. .blit
  9. .IS
  10. _sent

Suppresses

  1. InstanceOf
  2. ham
  3. laws
  4. Up
  5. h
  6. 袜
  7. down
  8. 询问
  9. uated
  10. haul