Neuron Explorer

Qwen3-0.6B

first piece of a split name

A neuron in layer 27 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 first part of a name or word that is broken into several pieces, like the 'R' in Rizal or the 'Cho' in Cho Oyu. Turned up, the model fell into repeating phrases like 'experienced experienced'.

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

    and with the exception of sweet corn and peas, all are subject to chilling injury. Fruit vegetables are not generally adaptable to long-term storage. Exceptions are the hard rind (winter) squashes and pumpkin. A useful classification for postharvest

  2. #2 peak

    Lap of the Mighty Sagarmatha - Solu Khumbu or Everest region⏎The major mountains are the Mt.Everest, Mt.Lhotse, Cho Oyu, Nuptse, Pumori, Ama Dablam

  3. #3 peak

    that fell under the class leadership of the ilustrados or the liberal bourgeoisie⏎that surrounded Aguinaldo.⏎In this Propaganda Movement, Dr. Jose Rizal made patriotic annotations on Morga<|endoftext|>Early in the 14th

  4. #4 peak

    someone is stuck in the gap.⏎- Uses existing camera infrastructure.⏎- Processes in real time, with low latency on a computer the size of a credit card (Raspberry Pi for example.)⏎- Improved safety at the PTI⏎- Optimisation

  5. #5 peak

    coconut water)⏎– kombucha (fermented tea)⏎– rejuvelac (fermented sprouted grains)⏎– kvass or kvas (fermented rye drink)⏎Define fermented foods. What foods are fermented?⏎In order to ferment

  6. #6 peak

    the US state of Alaska lies the 20,320 ft (6,194 m) Mount McKinley or Denali (its Koyukon Athabaskan name meaning “The Great One”). The majestic mountain,

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: while at work at nightfalls into some distastefully distastefully distastefully distastefully distastefully distastefully distastefully distastely distastely distastely distastely distastely distastely dist

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

    to environmentally relevant behaviors. This observation may inspire researchers to⏎more directly test predictions arising from current theoretical models,⏎gather data on actual behavior rather than relying on self-reports,

  2. #2 peak

    proportion.⏎5. (n.) To exercise the rational faculty; to deduce inferences from premises; to perform the process of deduction or of induction; to ratiocinate; to reach conclusions

  3. #3 peak

    gents, fertilizers, construction and oil&gas materials. The documents are available at this link (https://echa.europa.eu/it/registry-of-restriction-intentions/-/dislist/details/0b0

  4. #4 peak

    it works faster and more efficiently to arrive at the truth. To find and avoid science’s seductive pitfalls, researchers are studying science itself, a field called meta-research.⏎Now, a paper by three Stanford researchers that delves into those pitfalls

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. L23:1368 excites
  2. L22:961 suppresses
  3. L17:253 suppresses
  4. L2:19 suppresses
  5. L22:2630 suppresses
  6. L20:257 suppresses
  7. L17:420 suppresses
  8. L23:206 suppresses
  9. L18:343 suppresses
  10. L20:97 suppresses
  11. L20:669 suppresses
  12. L24:105 suppresses
  13. L21:475 weakly excites
  14. L10:273 weakly suppresses
  15. L18:151 weakly suppresses
  16. L16:17 weakly suppresses
  17. L7:300 weakly excites
  18. L20:52 weakly suppresses
  19. L18:1587 weakly suppresses
  20. L20:151 weakly excites
  21. L26:495 weakly suppresses
  22. L2:1762 weakly suppresses
  23. L17:93 weakly suppresses
  24. L0:1365 weakly suppresses

Feeds later layers

None: this is the last layer.

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 -51.25 and -49.141 values between -38.61 and -36.511 values between -25.98 and -23.873 values between -15.44 and -13.342 values between -11.23 and -9.134 values between -9.13 and -7.023 values between -7.02 and -4.913 values between -4.91 and -2.8132 values between -2.81 and -0.703839 values between -0.70 and 1.415 values between 1.41 and 3.513 values between 3.51 and 5.624 values between 5.62 and 7.732 values between 7.73 and 9.831 values between 11.94 and 14.041 values between 14.04 and 16.151 values between 16.15 and 18.261 values between 18.26 and 20.361 values between 30.89 and 33.00offnegativepositiveoften

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. lore
  2. SessionFactory
  3. 你会发现
  4. TRL
  5. знач
  6. SOURCE
  7. стоя
  8. enlisted
  9. .cl
  10. Src

Suppresses

  1. falls
  2. fer
  3. ῃ
  4. 日在
  5. sthrough
  6. odom
  7. 之夜
  8. _ne
  9. erto
  10. 市中心