word fragments inside longer words

What this neuron does

This neuron fires on a piece of a longer word that has been split apart, like the middle of 'Ajzen', 'irrevocable' or 'ISRO'. Turning it up made the model's writing drift into repetitive loops, so the effect is hard to read.

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

   975) Theory of Reasoned Action (TRA) and Ajzens (Ajzen 1985) extended Theory of Planned Behavior (TP

2. #2 peak

   round⏎- A photo of someone you love⏎- Something you love to play with<|endoftext|>The report’s focus is on perceptions of changes among pastoralists and agropastoralists in Borana, Ethiopia and Gao and Mopti,

3. #3 peak

   It can be at any time and for any reason. However, if the trust is an irrevocable living trust the Settlor does not have the authority to terminate the

4. #4 peak

   concern when considering the changing demographics.⏎The intention to perform a behavior is the most important factor to actually engaging in future behavior, according to Fishbein and Ajzens (Fishbein & Ajzen 1

5. #5 peak

   6 satellite into orbit to process information on climate, soil, seas and crop location.⏎The Spot 6 satellite is currently situated at Indian Space Research Organization (ISRO) facilities awaiting its scheduled launch in September.⏎Mexico also will have information delivered from

6. #6 peak

   in the pipeline, even if much of the energy they produce is destined for export.⏎On past trends, deforestation and soil erosion will follow, along with irrevocable changes to the course and character of untamed rivers, a quarter of

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

   be chromatically opposite, a very basic 3D stereoscopic effect is created.⏎Passive Technology (aka Polarised)⏎If you’ve been to an IMAX 3D cinema, then this is the technology you would have witnessed. It

2. #2 peak

   , which is life as usual for a Python programmer.⏎if not server_response['successful']:⏎etag = server_response['file_handle'].info().getheader('ETag')⏎xml_document = server_response['xml_document']<|endoftext|>Deepak Vashish

3. #3 peak

   ureau of Labor Statistics, U.S. Department of Labor, The Economics Daily, Demographics of Gulf War-era II veterans at https://www.bls.gov/opub/ted/2010/ted_201008

4. #4 peak

   artificial shapes, as is, unfortunately, too often the case.⏎C. PURPUREUS.—Purple Broom. Austria, 1792. Alow, spreading shrub, with long wiry shoots, clothed with neat trif

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.