end of document boundary token

What this neuron does

This neuron lights up at the invisible marker that signals one web page or document has ended and another begins. It is mostly quiet on ordinary words. Turning it up made the model's writing drift into unrelated, rambling topics.

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

   this video with your parents, friends, and/or adults and ask them if they can solve the problem as well⏎*Inspired by the movie Good Will Hunting<|endoftext|>Keats was one of the most life-loving people you could hope to meet,

2. #2 peak

   .⏎The source code (with examples) is here: hideinfo.cpp.⏎This software comes with no warrantee of any sort, use at your own risk!<|endoftext|>Mind Ransomware is one of those infections that are supposedly released for “educ

3. #3 peak

   1⏎By the end of Tuesday 16th please upload your piece ‘On D’ onto Moodle. Lots of dynamics and effects and different textures please!<|endoftext|>WOLFRAM SYSTEM MODELER⏎This

4. #4 peak

   00 training examples, you can choose to pick 100 random samples (batch size=100) for each EPOCH (Training cycle)<|endoftext|>|Origin:||co

5. #5 peak

   that is made terrible by our own mad attempt to interpret it as though it had⏎an underlying truth."⏎-- Umberto Eco⏎More information about the Python-list<|endoftext|>Jayne Davis: “I am not convinced in my area of research there are

6. #6 peak

   .⏎Download or subscribe to this show at https://twit.tv/shows/triangulation.⏎Thanks to Cachefly for the bandwidth for this show.<|endoftext|>The Scarlet Letter is an 1850 romantic work of fiction in a

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

   FLs). Installing one CFL saves $15 annually, according to the Rocky Mountain Institute, a nonprofit energy research group. Multiply that by the number of lamps and fixtures in your home, and watch the savings add up. Some people also prefer

2. #2 peak

   more likely.⏎Many scientists believe that the combination of a hereditary susceptibility to cancer and an environmental trigger, such as exposure to a carcinogen or being obese, puts individuals at risk of developing cancer, especially at an earlier adult age.<|endoftext|>Found

3. #3 peak

   . The wholesalers and distributors took those inflated orders at face value and, erring on the side of caution, ordered even more appliances. Those seemingly small decisions added up to big problems for General Electric.⏎Forrester’s system dynamics perspective has helped

4. #4 peak

   www.healthscout.com/template.asp?id=669525<|endoftext|>It is certain, in any case, that ignorance, allied with power, is the most ferocious enemy justice can have.⏎– James Baldwin⏎Four decades of neoliberal

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.