full stops closing a claim

What this neuron does

The neuron fires on the full stop or question mark that ends a sentence making a complete point, especially in persuasive or explanatory English, such as 'Either true or false.' Turning it up did not produce anything clearly related, mostly drifting into Chinese text and repetition.

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

   Non-cognitivists – moral statements are non-propositional and therefore subjective.⏎Propositional – deal with verifiable facts. Either true or false. We can legitimately ask ‘is it true?’⏎Non-propositional statements are

2. #2 peak

   day, the Government commissioned London’s sewer network, and one Thomas Crapper exploited the situation by patenting a range of flushing toilets.⏎And there we have it. For over 150-years the technology behind the flushing toilet has been faithfully

3. #3 peak

   in the future.⏎Our parenting advice is given as suggestions only. We recommend you also consult your healthcare provider, and urge you to contact them immediately

4. #4 peak

   But that’s neither here nor there. And kind of gross.)⏎It should be noted that the high diffusivity of caffeine is hardly a unique property, however. There are plenty of other compounds that have similar chemical backbone structures and will behave very

5. #5 peak

   so we don’t become rebellious and wayward.⏎We have two aspects. One is forgetful. It becomes wayward sometimes, absorbed with things that are

6. #6 peak

   ago, an ultra-conservative Floridian took his case to his state supreme court to fight for his right to wear a mask? Yep. Is this fake news? Nope.⏎So what gives? Why do so many people now philosophically insist that

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

   法、分析法来进行研究,探讨泰国学生在学习汉语助动词中的困难及偏误情况,分析他们产生偏误的原因。我们发现经过基于对比研究设计的汉语助动词课堂教学后,“

2. #2 peak

   东方在线 发布人：young⏎14. Energy Resources⏎China is rich in various energy resources.

3. #3 peak

   對 CNY 感興趣？⏎Early Currency in China⏎With a history of over 3000 years

4. #4 peak

   secretpassword';⏎mysql> GRANT ALL PRIVILEGES ON database_name.* TO user_name@localhost;⏎

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.