INDEX

Explanations

evaluation and qualities

The neuron fires on subjective evaluative or emphatic words and phrases that convey positive judgment or strong emphasis (e.g. really, totally, good, simple, relevant, relate).

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

of

-1.37

任何

-1.33

</i>

-1.23

 ANSWERS

-1.19

↩

-1.18

contentPane

-1.16

Lai

-1.13

uj

-1.12

 dtype

-1.11

熠

-1.10

POSITIVE LOGITS

Some

1.43

as

1.35

 Vereinigten

1.29

 Some

1.27

 there

1.24

it

1.23

there

1.23

álních

1.21

nitř

1.20

There

1.10

Activations Density 0.092%