"any" in questions and negatives

What this neuron does

This neuron reacts to the word "any" when it shows up in questions, negatives or "if you have any" phrases. It was strongly negative on the word "every" in a repeated company boilerplate line. Turning it up gave text that did not clearly connect to the idea.

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

   number of ways two blue triangles can⏎be fitted together?⏎Using different numbers of sticks, how many different triangles are⏎you able to make? Can you make any rules about the numbers of⏎sticks that make the most triangles?⏎Design an arrangement

2. #2 peak

   the corresponding digit. That means the last digit of the partial product must be in the same column as the digit of the second number. No product may begin with any additional zeros. If a particular digit is zero, the product has exactly one digit

3. #3 peak

   people may be trapped and are inaccessible to humans themselves. Lets hope nothing nefarious results from this; bugs are creepy enough as they are.⏎If you could create any type of insect robot, what would you make?<|endoftext|>- Historic Sites⏎It

4. #4 peak

   time. This lets you build a system that fits your energy needs and your budget.⏎I am still trying to make improvements to the design. So if you have any questions, problems or suggestions please leave a comment. I would really appreciate the feedback

5. #5 peak

   for the future.⏎How many inventions can you think of that were first made in the first decade of the first century AD? No, I can’t think of any either, and not just because it was a long time ago. There

6. #6 peak

   any link between the use of classical psychedelics and the risk of developing mental health problems

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.

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

   more about Healthwise, visit Healthwise.org.⏎© 1995-2016 Healthwise, Incorporated. Healthwise, Healthwise for every health decision, and the Healthwise logo are trademarks of Healthwise, Incorporated.

2. #2 peak

   wise⏎To learn more visit Healthwise.org⏎© 1995-2012 Healthwise, Incorporated. Healthwise, Healthwise for every health decision, and the Healthwise logo are trademarks of Healthwise, Incorporated.⏎

3. #3 peak

   wise⏎To learn more visit Healthwise.org⏎© 1995-2012 Healthwise, Incorporated. Healthwise, Healthwise for every health decision, and the Healthwise logo are trademarks of Healthwise, Incorporated.

4. #4 peak

   .⏎To learn more, visit Healthwise.org⏎© 1995-2015 Healthwise, Incorporated. Healthwise, Healthwise for every health decision, and the Healthwise logo are trademarks of Healthwise, Incorporated.

Which neurons in earlier layers feed this one, and which later ones it feeds. Amber links excite, blue links suppress; longer bars are stronger.

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.

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.