INDEX

Explanations

everyone and everything

The neuron responds to “every-” words used as universal quantifiers (e.g. “everyone,” “everything,” etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

↵↵

-2.23

or

-2.02

 restful

-1.99

 inconspicuous

-1.91

</i>

-1.88

 häufig

-1.77

↵

-1.76

\&

-1.72

 aspirations

-1.71

高い

-1.70

POSITIVE LOGITS

乄

2.25

 suí

2.23

2.08



2.08

緜

2.06

])->

2.06

轳

2.05

2.03

na

1.98

()

1.98

Activations Density 0.019%