INDEX

Explanations

social or physical context involving others

The neuron responds to longer, less common content words (i.e. relatively rare, multi‐syllable vocabulary) in the text.

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

0.72

for

0.55

0.54

0.53

 from

0.52

0.50

to

0.50

as

0.49

POSITIVE LOGITS

<unused547>

0.61

ᒍ

0.60

 अभिकारक

0.59

 ماحول

0.56

 Kommunikation

0.55

 CtApp

0.55

<unused589>

0.54

<unused1741>

0.54

<unused2132>

0.54

<unused1768>

0.54

Activations Density 0.000%