Neuron Explorer

Qwen3-0.6B

"by" introducing a deadline or means

A neuron in layer 6 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 fires on the word "by" when it introduces a deadline ("by 2030") or the way something is done ("by venturing into space", "by covering with tar"). Turning it up did not produce a clear change in the writing.

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

    at creating new, green jobs and investments in frontline communities bearing the brunt of environmental injustice—they offer a blueprint for both cutting U.S. carbon emissions in half by 2030 and doing so equitably.⏎But neither plan explicitly gr

  2. #2 peak

    up to seven astronauts.⏎China embarked on its manned space exploration odyssey in 2003 when Yang Liwei, a former fighter pilot, made history by venturing into space aboard the Shenz

  3. #3 peak

    importance of embracing the natural world.⏎Additionally, [Theme 2] is explored in the poem. Oliver delves into the theme of [Theme 2] by [Provide an analysis of how the theme is portrayed in the poem].⏎Use of

  4. #4 peak

    to the hub of the vehicle completely, or debris or corrosion may be between the hub and the rotor, causing this runout. A technician will correct runout by completely cleaning the surfaces where the hub and rotor meet. The rotor may also need

  5. #5 peak

    facilities. Nigeria has pledged its commitment to meeting Sustainable Development Goal (SDG) 6, which is to “ensure access to water and sanitation for all” by 2030. But the National Longitudinal Phone Survey (NL

  6. #6 peak

    19, he moved<|endoftext|>A team of researchers from the University of California have developed an optical processor designed to offer massive parallelization for deep-learning workloads — by giving a deep-learning workload the job of designing it.⏎"Large-scale linear operations

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: The first time, I have been to a place I have been to a place I have been to a place I have been to a place I have been to a place I have been to a place I have been to a place I have

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

    ?" Butte said. "It's a simple question that no one had answered in this way."⏎In their paper, published in the current issue of the Public Library of Science ONE, they say there's an established link between Type 1 diabetes

  2. #2 peak

    brochure is one response to a need identified by people with disabilities. The way we portray people with disabilities and our attitudes toward them are critical to their future...and to ours.⏎Disability vs. Handicap⏎A disability is a condition caused by

  3. #3 peak

    . The RIFA is widespread, but Florida is also home to a less common species known as the native or tropical fire ant. Neither of the species takes kindly to being stepped on, and they respond to such unwarranted behavior by attacking merciless

  4. #4 peak

    individuals and their cultures could impede pharmaceutical care. Incongruent beliefs and expectations between the patient and pharmacist could lead to misunderstandings, confusion, and ultimately to drug misadventures. Models and frameworks have been developed that provide descriptions of

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. L4:858 excites
  2. L4:1394 weakly excites
  3. L4:1711 weakly suppresses
  4. L5:2641 weakly excites unclear: scattered word fragments
  5. L0:872 weakly excites
  6. L4:809 weakly suppresses
  7. L3:381 weakly suppresses 'for' in research institute names
  8. L4:57 weakly suppresses
  9. L5:1545 weakly excites
  10. L4:1460 weakly suppresses
  11. L4:2893 weakly excites the word 'means' after a subject
  12. L0:2561 weakly suppresses
  13. L5:880 weakly excites
  14. L3:545 weakly excites
  15. L3:1579 weakly suppresses
  16. L1:2703 weakly excites
  17. L2:1477 weakly suppresses
  18. L5:2392 weakly suppresses
  19. L2:2018 weakly excites
  20. L4:486 weakly excites
  21. L4:1093 weakly excites
  22. L1:578 weakly excites
  23. L2:1687 weakly excites
  24. L4:2363 weakly excites

Feeds later layers

  1. L8:94 strongly excites
  2. L8:369 strongly suppresses 'of' after 'learned' or 'because'
  3. L9:226 strongly excites 'to' starting an infinitive
  4. L16:1153 excites
  5. L15:1176 excites
  6. L14:659 excites
  7. L8:723 excites
  8. L18:472 excites
  9. L7:138 excites 'to' after verbs like prove or time
  10. L17:472 excites
  11. L17:1592 excites
  12. L16:559 excites weak, mixed: verbs and adverbs in claims
  13. L7:681 excites 'from' after prevent or stop
  14. L7:15 excites
  15. L19:385 suppresses
  16. L19:652 excites
  17. L15:1234 excites
  18. L17:187 excites
  19. L19:617 excites
  20. L18:715 excites
  21. L22:200 excites
  22. L16:1114 excites
  23. L13:909 excites 'to' before a verb after a noun
  24. L18:452 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.

1 values between -1.39 and -1.271 values between -1.27 and -1.151 values between -1.03 and -0.922 values between -0.92 and -0.803 values between -0.80 and -0.687 values between -0.68 and -0.5622 values between -0.56 and -0.4461 values between -0.44 and -0.32242 values between -0.32 and -0.21674 values between -0.21 and -0.091576 values between -0.09 and 0.03940 values between 0.03 and 0.15271 values between 0.15 and 0.2760 values between 0.27 and 0.3912 values between 0.39 and 0.5015 values between 0.50 and 0.623 values between 0.62 and 0.744 values between 0.74 and 0.862 values between 0.86 and 0.981 values between 1.10 and 1.212 values between 1.21 and 1.331 values between 1.33 and 1.451 values between 1.45 and 1.571 values between 1.69 and 1.802 values between 1.92 and 2.041 values between 2.40 and 2.512 values between 3.22 and 3.34offnegativepositiveoften

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. 把她
  2. dig
  3. 让她
  4. alte
  5. イラ
  6. 对她
  7. 让消费者
  8. 把这个
  9. digitally
  10. 创造出

Suppresses

  1. izen
  2. urance
  3. concern
  4. ustr
  5. consistency
  6. lemen
  7. .toList
  8. 련
  9. usion
  10. ounded