INDEX

Explanations

relationships with

The neuron consistently lights up on short relational/function words and particles (e.g. “in,” “of,” “up,” “what,” “out”) that connect or frame actions in a sentence.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 самому

0.70

 mình

0.67

 ਆਪਣੇ

0.63

自己在

0.62

 خود

0.62

 اپنے

0.61

 নিজেকে

0.61

 বলেছি

0.61

 আছি

0.60

 스스로

0.59

POSITIVE LOGITS

us

1.62

 conmigo

1.52

my

1.48

跟我

1.46

和我

1.42

 मुझसे

1.38

 मुझ

1.37

 нас

1.32

me

1.31

 আমাকে

1.30

Activations Density 0.623%