INDEX

Explanations

inimitable sense

The neuron fires on content words that participate in fixed two‐word collocations (e.g. “hands on,” “close to,” “fact sheet,” “inimitable sense”).

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

by

-2.27

Anzahl

-1.52

%");

-1.48

Ẳ

-1.45

趼

-1.45

Cdb

-1.44

 analogy

-1.40

 afgelopen

-1.38

čený

-1.38

woordig

-1.38

POSITIVE LOGITS

1.64

чие

1.56

 prüfen

1.55

 they

1.44

เรา

1.43

1.38

 musíte

1.33

itola

1.30

Activations Density 0.036%