'for' after study resource nouns
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. 10 of Qwen3-0.6B. Its strongest snippets are very consistent: they mostly 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 responds to the word 'for' when it introduces who something is meant for, like 'Notes for the students' or 'solutions for Class 10'. Its negative side picks up words after verbs of purpose, such as 'make' or 'achieve'. Turning it up did not produce a clear theme in the written text.
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 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.
-
⏎7 letter words⏎6 letter words⏎5 letter words⏎4 letter words⏎3 letter words⏎2 letter words⏎Images for ZEATINSLoading...⏎SCRABBLE is the registered trademark of Hasbro
-
& Grow: all you need to know to get those veg plants off to a roaring start!⏎1. WHAT should I grow
-
Exercise 6.5⏎In NCERT Solutions for Class 10 Maths Chapter 6 Exercise 6.5, we have
-
Board Class 10 Science Chapter 4 Environmental management questions answers. Our expert team have gone through this chapter and prepared Notes for the students who are facing difficulties in summarizing of this chapter. We assure you
-
abstract concepts, such as friendship, or concrete concepts, such as the food chain.⏎Common Core Standards⏎This resource has been aligned to the Common Core State Standards for states in which they have been adopted. If a state does not appear in the
-
here. There are several questions in this exercise. Our subject expert teachers have solved each of these questions in a step by step manner. All these solutions are beneficial for the students of class 10 appearing for the board exams. Students can easily
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: I need to be able to get the right answer for the question of the day's example... I need to be able to get the right answer for the question of the day's example... I need to be able to get the right answer
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
of the Planet is presented with support provided by the National Science Foundation. I’m Jim Metzner.⏎Around the world, mangroves are being destroyed to make room
-
because of His needs. He did it illustrate the importance of Christ’s blood to man’s limited human understanding. It also revealed the immense sacrifice He would make for us.⏎Old Testament man had only one hope beyond the grave; this was God
-
10, in the 100 metropolitan areas with the largest African American populations, 62.5 percent of blacks would have had to move to achieve full black–white integration.⏎- Unemployment rates for African Americans have been far
-
, and professors were lenient when students turned in papers late, because everyone was piecing together so many jobs to get by. “We learned the value of hard work,” the former senior class president
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
- L9:1890 excites
- L6:2000 suppresses
- L7:1584 suppresses
- L9:583 suppresses
- L5:1890 excites
- L4:1020 suppresses
- L6:1551 suppresses
- L6:2120 suppresses
- L6:2738 weakly suppresses
- L7:989 weakly suppresses
- L9:2500 weakly suppresses
- L3:2433 weakly suppresses
- L7:928 weakly excites
- L6:2897 weakly suppresses
- L7:530 weakly suppresses
- L3:427 weakly excites
- L6:2880 weakly excites
- L4:1460 weakly suppresses
- L1:2346 weakly suppresses
- L9:817 weakly suppresses
- L5:801 weakly suppresses
- L5:2464 weakly suppresses
- L9:796 weakly suppresses
- L7:957 weakly suppresses
Feeds later layers
- L12:2151 excites
- L17:2856 suppresses
- L13:775 suppresses
- L14:2626 suppresses
- L12:2548 suppresses
- L14:2549 suppresses
- L12:2436 excites
- L12:2850 suppresses
- L12:2212 weakly suppresses 'to' or 'for' opening a purpose phrase
- L12:2685 weakly suppresses
- L14:785 weakly suppresses
- L12:2793 weakly suppresses
- L15:708 weakly suppresses
- L12:2669 weakly suppresses
- L15:1139 weakly suppresses
- L11:2117 weakly suppresses
- L21:2392 weakly excites
- L12:2402 weakly suppresses
- L20:1623 weakly suppresses
- L18:658 weakly suppresses
- L14:289 weakly suppresses
- L15:513 weakly suppresses
- L15:582 weakly excites
- L13:1375 weakly suppresses
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
- 直升
- olland
- hiro
- ishes
- ewis
- 萋
- isers
- 翅
- ogh
- NCY
Suppresses
- 的背后
- prevailing
- 掩
- 追求
- strument
- running
- ( ⏎
- .getLogger
- 达成
- 回答