unclear: first and boundary tokens

What this neuron does

This neuron's strongest values fall on the very first word of a text chunk or on the special end-of-text marker, which suggests it reacts to position rather than to meaning. The extreme values are unusual and the steering output is mostly repetition, so the purpose is unclear.

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

   The quest to achieve the impossible fuels creativity, spawns new fields of inquiry, illuminates

2. #2 peak

   The studio probably had little choice but to create a form letter.⏎I don’t think

3. #3 peak

   The recipient previously opened the message⏎Still, also the recall won’t work If a

4. #4 peak

   The Negro Speaks of Rivers,” written the summer after he graduated from high school in

5. #5 peak

   The unused funds — called Universal Service and Access Funds (USAFs) — are

6. #6 peak

   The review of 20 previous studies found that people who were given dark chocolate or

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

   Norman cathedral. It is the seat of the Bishop of Coutances and Avranches,

2. #2 peak

   )<|endoftext|>Trophic state and seasonal dynamics of phytoplankton communities in two sand-p

3. #3 peak

   <|endoftext|>What is Cancer? What Causes Cancer?⏎Cancer is a class of diseases characterized by

4. #4 peak

   brain<|endoftext|>Harm Is in the Eye of the Beholder⏎Mark Davis is the De

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.