INDEX

Explanations

hands and arms

The neuron strongly activates on mentions of the body parts “hands” (and to a lesser extent “feet”), i.e. occurrences of those limb words.

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

-1.43

 són

-1.37

-1.31

-1.30

螄

-1.25

 milliers

-1.24

of

-1.24

-1.22

 instead

-1.20

 echte

-1.19

POSITIVE LOGITS



1.76

媖

1.63

 Such

1.61

࿚

1.56

鐧

1.54

 éditoriale

1.52

 contagio

1.52

”)

1.49

 What

1.49

 Again

1.48

Activations Density 0.091%