INDEX

Explanations

child elements or items

The neuron activates on numeric literals (especially floating-point numbers) enclosed in parentheses.

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

 myös

-1.87

 också

-1.82

 ſont

-1.73

kuu

-1.63

toxicity

-1.63

该如何

-1.61

いいのか

-1.52

鸡

-1.52

 oogen

-1.52

 parha

-1.51

POSITIVE LOGITS

1.38

—

1.29

at

1.25

↵↵

1.23

and

1.22

teh

1.22

 más

1.20

一件

1.20

ociação

1.20

Activations Density 0.063%