INDEX

Explanations

parentheses followed by specific words

function calls in code

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

 failings

0.52

.’

0.48

!’

0.46

।’

0.44

-‘

0.44

 prohibitions

0.44

 shortcomings

0.44

}']

0.43

′

0.43

,’

0.42

POSITIVE LOGITS

Hello

0.98

----------------

0.93

{\"

0.92

================

0.88

There

0.88

This

0.87

hello

0.85

<html>

0.84

Here

0.84

=============

0.84

Activations Density 0.003%