INDEX

Explanations

positive or negative sentiment about things

The neuron specializes in spotting strongly evaluative sentiment words (e.g. loved, adored, despised).

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

to

-1.75

になるので

-1.44

 nunca

-1.38

èvent

-1.38

 propiet

-1.37

 colección

-1.37

</u>

-1.37

݈

-1.34

 TestBed

-1.34

те

-1.32

POSITIVE LOGITS

by

2.05

if

1.70

 всеми

1.56

in

1.48

 when

1.48

 with

1.39

</h2>

1.36

''

1.35

 czerwca

1.34

졌

1.32

Activations Density 0.108%