math word problems and worksheets
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 mixed: they peak on different tokens. 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 math teaching text, especially words like 'problem', 'word problems' and 'numbers' in worksheets and lessons. Turning it up pushed the writing toward problems and numbers, though the text got muddled. Turning it down made it repeat itself.
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.
-
email@example.com.<|endoftext|>Table of Contents :⏎Top Suggestions Senior Kindergarten Writting Worksheet :⏎Senior Kindergarten Writting Worksheet In this worksheet your students will practice their handwriting skills by tracing the
-
39 = 611 candles.⏎Using a bar diagram⏎In the bar diagram, the relationship is clear. This is a simple, one-step problem.⏎650 – 39 = 611 candles sold.⏎
-
ing (2-digits & 3-digits)⏎✔ Subtraction with and without regrouping (2-digits & 3-digits)⏎✔ Word Problems: addition, subtraction, and multiplication⏎✔ Addition Fact Fluency⏎✔ Sub
-
know multiplying fractions do not require the same denominators. So click and give it a try. It's easier than you think.⏎Fractions - Subtracting Mixed Numbers⏎Do you wonder how to subtract mixed numbers? Do you know how to handle
-
so PLEASE, PLEASE leave a comment if you LOVE your math series! I'd love to check it out!⏎I like the tricks (9s trick, Doubles, Doubles Plus 1) that Saxon uses to teach facts, but
-
How many candles did they sell?⏎This is much like a 2nd-grade problem, but with larger numbers. The main difficulty is to convert the common-sense
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: "This is a bird's eye view. It's the first thing you see. You take the first step. You take the first step. You take the first step. You take the first step. You take the first step. You
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
time should focus on four critical areas: (1) developing understanding of and applying proportional relationships; (2) developing understanding of operations with rational numbers and working with expressions and linear equations; (3) solving problems involving
-
draw and write to show their mathematical thinking.⏎Contents 4-8 Counting and Cardinality Task Cards (CC) 9-13 Numbers and Operations in Base Ten Task Cards (NBT) 14-18 Operations and
-
before returning it to the shelf. Through time, the child refines their grip of the paintbrush. What a gentle, inviting way to learn how to later hold a pencil for writing!⏎Why are art activities important?⏎Dr. Montessori
-
language. Octave has extensive tools for solving common numerical linear algebra problems, finding the roots of nonlinear equations, functions written in Octave's own
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
- L7:573 suppresses
- L7:2349 suppresses
- L6:1552 suppresses
- L6:1840 weakly excites
- L9:1453 weakly suppresses
- L7:2093 weakly suppresses
- L9:2665 weakly excites
- L9:1547 weakly suppresses
- L7:909 weakly excites
- L8:1709 weakly suppresses
- L6:987 weakly suppresses
- L2:1307 weakly suppresses
- L7:494 weakly excites
- L6:2406 weakly excites
- L6:1073 weakly excites
- L9:2930 weakly suppresses
- L6:2613 weakly excites
- L6:2362 weakly suppresses
- L2:2369 weakly excites
- L8:950 weakly suppresses
- L3:2474 weakly excites
- L7:121 weakly suppresses
- L8:2344 weakly excites
- L7:611 weakly suppresses
Feeds later layers
- L21:2366 excites
- L13:2592 excites
- L11:924 excites
- L11:1119 excites
- L11:1295 excites
- L22:2870 weakly suppresses
- L12:777 weakly excites
- L17:1656 weakly excites
- L11:3044 weakly excites
- L11:661 weakly excites
- L23:269 weakly excites
- L18:2371 weakly excites
- L21:410 weakly excites
- L18:387 weakly suppresses
- L13:107 weakly excites
- L17:2349 weakly excites
- L15:2634 weakly excites
- L18:140 weakly excites
- L11:1183 weakly suppresses
- L17:1393 weakly suppresses
- L21:1363 weakly excites
- L21:2739 weakly excites
- L11:1661 weakly excites
- L22:2277 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
- astos
- 題
- _cases
- casos
- 题目
- ést
- äsent
- cases
- -str
- exercises
Suppresses
- 六个月
- (IT
- 人流
- [,
- 当晚
- PRI
- 心跳
- offending
- "],["
- 排名第一