zeros inside large round numbers

What this neuron does

This neuron fires on a zero in large round numbers like 10,000 or 3,000 people, where the zero is part of the number's size. Turning it down made the model write a long string of zeros, which matches what it responds to.

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 activation was highest. Amber marks positive activation; the outlined token is the peak. Hover a token for its strength.

1. #1 peak

   0 years ago.⏎Food used to be obtained through hunting and gathering. The paleo diet restricts foods that became common when agriculture appeared about 10,000 years ago

2. #2 peak

   scientific curriculum that included linguistics, arts and crafts, medicine, logic, and spirituality. People in India continue to practice two ancient systems developed over 2,000 years ago to maintain health: Ayurvedic medicine, described by the

3. #3 peak

   ense with the stored collection of models and gave the Smithsonian Institution the opportunity to collect any models it wanted. Today, the Museum’s collection exceeds 10,000 patent

4. #4 peak

   bitaceae family. Here are some Butternut Squash Health Benefits.⏎Squash is among the oldest known crops dating back 10,000 years to Mexico and America. The name squash, which means uncooked or

5. #5 peak

   the Gas Giant Caligine. Before the Great Crusade, Caligine II was a bustling mining world that at its peak employed more than 3,000,000 people. This population lived in cramped conditions while reinforced

6. #6 peak

   more than 400 varieties of biscuits and employed more than 5,000 people.⏎Huntley & Palmers employed the latest technology but their

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

   ória [online]. 2006, vol.25, n.1, pp. 44-69. ISSN 1980-4369. http://dx.doi.org/10

2. #2 peak

   , MA: WGBH Media Library & Archives. Retrieved from http://openvault.wgbh.org/catalog/V_4E4EE3EAD24043E2BC0B972CD24B1

3. #3 peak

   5, it would imply the exit of 166.5 million individuals from poverty. In reality, the number of poor in 2004-05 turned out to be 302 million, implying that India had

4. #4 peak

   VECE20|⏎When requesting a correction, please mention this item's handle: RePEc:taf:jeduce:v:42:y:2011:i:2:p

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.