INDEX

Explanations

live in, preset preferences, csv data, zip file

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

in

-2.05

at

-1.85

on

-1.73

 from

-1.67

for

-1.61

 like

-1.47

 beberapa

-1.42

is

-1.41

 across

-1.41

 when

-1.41

POSITIVE LOGITS

mathrm

2.05

𝐌

1.52

medim

1.51

miyorum

1.48

↵

1.44



1.44

𝐓

1.43

 květ

1.43

²,

1.41

})$,

1.38

Activations Density 0.161%