Neuron Explorer

Qwen3-0.6B

stem cell terms in biology text

A neuron in layer 8 of Qwen3-0.6B. Its strongest snippets are very consistent: they mostly peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron lights up on the words 'stem' and 'self-renew' when they appear in articles about stem cells and cancer. When it was turned up, the written text drifted to unrelated topics, so the steering results don't show a clear effect.

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

    is deregulated in many types of human malignancies, including cancers of kidney, liver, stomach, breast, and brain. Normally, only stem cells and progenitor cells express MET, which allows these cells to grow invasively in order to

  2. #2 peak

    , and hydrangea. Rejuvenation pruning is best done in late winter or early spring.<|endoftext|>Neural stem cells (NSCs) can self-renew and differentiate into neurons and glia. Transplanted NS

  3. #3 peak

    of U.S. scientists has identified phosphorylated signalling proteins in human embryonic stem cells (hESCs) studying which may help understand the mechanisms that determine whether stem cells divide or differentiate

  4. #4 peak

    was poorly understood.⏎According to them, identification of these phosphorylation sites provides insights into known and novel hESC signalling pathways and highlights signalling mechanisms that influence self-renewal and differentiation.⏎"This research will be a big boost for stem cell scientists.

  5. #5 peak

    over remarkably long distances, and glia. Grafts of cultured human NSCs expressing GFP resulted in similar findings. Thus, methods are defined for improving neural stem cell grafting, survival and analysis of in vivo findings.⏎21 Related Jo

  6. #6 peak

    genetics at Harvard Medical School (HMS). "It also shows, from one lung cancer to another, the cancer stem cells are not the same."⏎Cancer stem cells are a subset of cancer cells believed to elude conventional treatments and eventually regenerate

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: which is the way to spread this, so 100 is the coefficient to be used when no decimal is involved. The formula used in this case, which is to be used when no decimal is present, so the formula for

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

    only music production and records selling, but also touring and merchandising. Artists themselves are looking for new ways to make money,

  2. #2 peak

    his songs always contain the words ‘Guruguha’ towards the end.⏎In his late life, he toured South India extensively and composed songs on the various deities of the religious pantheon

  3. #3 peak

    computer science teacher Myra Deister of Sunny Hills High in Fullerton didn’t learn computer programming in high school.⏎Neither did most of the professionals who come to speak to her classes about their careers. Across the country, just 10 percent

  4. #4 peak

    with the procedure now correctly termed Mohs Micrographic Surgery.⏎After the removal of the visible portion of the tumor by excision or curettage (debulking), there are two basic steps to each Mohs Micrographic Surgery stage. First

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. L7:2267 suppresses
  2. L6:1916 weakly excites "stem" in stem cell phrases
  3. L5:1395 weakly excites
  4. L4:2759 weakly excites
  5. L4:2988 weakly excites
  6. L0:2539 weakly suppresses
  7. L4:1053 weakly excites
  8. L6:1924 weakly suppresses
  9. L7:63 weakly excites
  10. L4:1050 weakly excites
  11. L4:3038 weakly excites
  12. L3:2428 weakly suppresses
  13. L2:1688 weakly suppresses 'off of' and 'of' after 'off'
  14. L4:2645 weakly suppresses
  15. L0:2137 weakly excites
  16. L7:2185 weakly excites
  17. L6:2411 weakly suppresses
  18. L6:1142 weakly excites
  19. L7:191 weakly excites
  20. L7:2149 weakly suppresses
  21. L7:1838 weakly excites
  22. L7:49 weakly suppresses
  23. L0:106 weakly suppresses unclear: 'if' opening a statement, 'local'
  24. L1:1462 weakly excites

Feeds later layers

  1. L10:1312 excites
  2. L9:0 excites
  3. L9:1413 weakly excites
  4. L20:2606 weakly excites
  5. L22:2886 weakly excites
  6. L11:1519 weakly excites
  7. L10:1064 weakly excites
  8. L16:845 weakly excites
  9. L9:2292 weakly excites
  10. L9:1553 weakly excites
  11. L20:2878 weakly excites
  12. L19:2442 weakly excites
  13. L9:1679 weakly suppresses
  14. L11:666 weakly excites
  15. L19:387 weakly suppresses
  16. L22:1521 weakly excites
  17. L9:357 weakly excites
  18. L10:623 weakly excites
  19. L16:2931 weakly excites "all" meaning every level or group
  20. L12:1227 weakly excites
  21. L25:2064 weakly excites
  22. L24:1810 weakly suppresses
  23. L13:3009 weakly suppresses
  24. L11:906 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.

2 values between -1.03 and -0.941 values between -0.85 and -0.763 values between -0.76 and -0.673 values between -0.67 and -0.585 values between -0.58 and -0.499 values between -0.49 and -0.4038 values between -0.40 and -0.31129 values between -0.31 and -0.22429 values between -0.22 and -0.12886 values between -0.12 and -0.031220 values between -0.03 and 0.06678 values between 0.06 and 0.15317 values between 0.15 and 0.24126 values between 0.24 and 0.3332 values between 0.33 and 0.428 values between 0.42 and 0.514 values between 0.51 and 0.603 values between 0.60 and 0.693 values between 0.78 and 0.872 values between 0.96 and 1.051 values between 1.05 and 1.141 values between 1.14 and 1.232 values between 1.32 and 1.423 values between 1.51 and 1.601 values between 1.69 and 1.781 values between 1.87 and 1.961 values between 2.50 and 2.59offnegativepositiveoften

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. subclass
  2. omorphic
  3. .imgur
  4. 细胞
  5. .createCell
  6. filtr
  7. subclass
  8. feb
  9. _cells
  10. 층

Suppresses

  1. ounds
  2. linky
  3. laces
  4. Scandinavian
  5. Unknown
  6. 隍
  7. igator
  8. Unknown
  9. undisclosed
  10. 地