parenthesis after prophet or Allah names
A neuronOne of thousands of tiny dials inside each layer. It reads the text so far and produces a number, its activation, that nudges what the model writes next. in layerOne step in the stack the text passes through. Qwen3-0.6B has 28 layers, each with 3,072 neurons. 11 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featuredA neuron picked for the full treatment: steering outputs and a written explanation. About 3,000 of the 86,000 neurons are featured., so it has steering outputs.
What this neuron does
This neuron fires on the opening bracket that follows honorific names, like 'Allah (swt)' or 'the Prophet (peace be upon him)'. Those are mostly in Islamic religious writing. Steering did not show a clear theme.
Written by Claude from this neuron's activations and steering outputs. Confidence: high. Think it is wrong? Flag it (opens a prefilled issue on GitHub).
Fires most on
The text windows where this neuron's activationThe number a neuron produces for a token. Large positive means it fired hard; negative means it pushed the other way. was highest. Amber marks positive activation; the outlined tokenA piece of text the model reads at once: often a word, sometimes part of a word or a punctuation mark. is the peak. Hover a token for its strength.
-
certainly supports our point of view that in the grief of a departure, wailing is a natural habit. Wailing or weeping is a natural instinct hence Allah (swt) has
-
2:228)⏎Through the beautiful discourse of the Holy Quran and Sunnah, Muslims come to learn about the various rights that Allah (SWT) has enjoined upon His creation – in a practical and detailed manner
-
another act of worship is regarded as general du’aa’; in this case there is nothing wrong with raising the hands. It was narrated that the Prophet (peace and blessings of Allaah be upon him) spoke of a man who raised
-
the Holy Prophet (upon whom be peace) and its object to remove his anxiety and distress, which
-
وان مقاله [English]⏎نویسنده [English]چکیده [English]⏎The author in this article has chosen all Imam Ali’s (A.S) speeches in Nahj-Albalagheh which are specifically about
-
things, Allah (SWT) has made men the protectors and guardians of women – it is
SteeringAdding a push along a neuron while the model writes, to see how its output changes. Turning it up or down.
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.
Here is something I have been thinking about lately: the need for a plan. I need to get the people I know, the ones who are in a business, maybe the finance, the people who have a small business, maybe a home, maybe a home, maybe a home, and
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
into the mix. Excitingly, these histones, when expressed in E. coli, bind the genome nearly<|endoftext|>Definition of Nupe in English:⏎noun (plural same or Nupes)⏎2 [mass noun] The Benue-C
-
review worksheet, students are given sentences and must find the mistake. Students correct the mistakes when they find them using their knowledge of verb tense.⏎3 Views 28 Downloads<|endoftext|>Blank Electoral College Map 201
-
15(2):57-61 DOI:http://dx.doi.org/10.4314/aas.v15i2.5⏎© 2018 Author. This work is licensed under the
-
systems.⏎What is Lyra's demon?⏎lyra's demon, pantalaimon /ˌpæntəˈlaɪmən/, is his most beloved companion, whom he calls “
Connections
Show on mapWhich neuronsOne of thousands of tiny dials inside each layer. It reads the text so far and produces a number, its activation, that nudges what the model writes next. in earlier layersOne step in the stack the text passes through. Qwen3-0.6B has 28 layers, each with 3,072 neurons. feed this one, and which later ones it feeds. Amber links excite, blue links suppress; longer bars are stronger.
Fed by earlier layers
- L10:1785 suppresses
- L10:252 excites
- L10:510 excites
- L10:500 excites
- L8:1728 suppresses
- L10:1632 suppresses
- L8:790 excites
- L9:1593 excites
- L10:1706 weakly suppresses
- L8:756 weakly suppresses
- L10:2810 weakly suppresses
- L9:776 weakly suppresses
- L9:2057 weakly excites
- L1:1817 weakly excites
- L8:251 weakly suppresses
- L10:1043 weakly suppresses
- L9:308 weakly excites
- L10:1013 weakly suppresses
- L4:953 weakly suppresses
- L10:2036 weakly suppresses
- L7:2038 weakly excites
- L7:575 weakly suppresses
- L10:976 weakly excites
- L9:3010 weakly suppresses
Feeds later layers
- L15:943 excites
- L15:988 excites
- L12:1598 excites
- L17:2354 excites
- L12:2700 excites
- L13:1522 excites
- L13:283 excites
- L13:8 suppresses
- L12:120 suppresses
- L12:787 excites
- L13:2352 excites
- L13:1900 excites
- L14:666 suppresses
- L13:1748 weakly excites
- L15:420 weakly excites
- L12:1826 weakly excites
- L15:30 weakly excites
- L14:2032 weakly excites
- L12:507 weakly excites
- L13:2450 weakly excites
- L13:2847 weakly excites
- L13:1167 weakly excites
- L16:1226 weakly excites
- L17:2402 weakly excites
Activation histogramA chart of how often a neuron takes each activation value.
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.
Logit lensReading which output tokens a neuron pushes toward or away from, straight off its output weights, before the later layers have their say.
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 layersOne step in the stack the text passes through. Qwen3-0.6B has 28 layers, each with 3,072 neurons. instead.
Promotes
- tử
- esse
- esser
- سابق
- _stderr
- orget
- costa
- Administrator
- 謝
- essen
Suppresses
- ICENSE
- GEST
- .Line
- built
- -'
- onden
- ıyor
- very
- easily
- 深层