unclear: verbs after a noun or 'that'

What this neuron does

This neuron is hard to pin down. It fires on verbs such as 'harvest', 'used', 'burn' and 'molting' that come right after a noun or 'that', but the pattern is loose. Steering did not produce a clear theme.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

   . As a result, farmers will be able to significantly reduce pesticide applications and harvest bigger yields. The crop is also expected to support the nation's economic development and

2. #2 peak

   structure is similar to that of the ___2___ and national governments, with an elected mayor as chief of the ___3___ and an elected council that ___4___ the various neighborhoods forming the ___5___ . The ___6___ appoints heads

3. #3 peak

   For cantaloupe, the principal harvest indices are surface color and the development of the abscission zone.⏎Most fruit vegetables are harvested by hand. Some harvest aids may be used, including pickup machines and conveyors

4. #4 peak

   not found anywhere else, and now rare even in Jordan; the Christian Bedouins.⏎Many of the traditional houses of Hmoud were built on caves that used to be a strong attraction for settling this site, as they provided early shelters prior

5. #5 peak

   be extremely stingy with neutrons. The coolant moderator combination had to be one that did not absorb neutrons to any great extent. To make natural uranium “burn” every available neutron has to find its way into fuel material. Using a neutron

6. #6 peak

   izing the cities.⏎Designing the board, deciding how to connect the cities⏎Creating their board based on having 6 players⏎Setting up the game to playtest it⏎“The city defence changes when

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

   while loop keeps running while the value of i is equal to or greater than 0.⏎- Inside the loop, we are printing the element for index i and decrementing the value of i by 1.⏎If you run this program, it

2. #2 peak

   ⏎Create sequnce 1 4 9 16 25.........⏎Basics: Divide integers to get integer outputs in all cases⏎Determine value using Look-up table method⏎Create Volcano (or Atoll) mart

3. #3 peak

   plane is still on the ground, you design and develop measuring devices that take temperature, noise, altitude, navigation, and maneuverability readings. Then you hard-wire the devices into place, and send the Test Pilot into the clouds. When he

4. #4 peak

   progress over the years!⏎- Color and cut out the pieces.⏎- Roll the large triangular piece and use the tab to glue it into a cone⏎- Glue the nose onto the face using a dot of glue on the nose (this allows

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.