INDEX

Explanations

numbers followed by commas or letters

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

and

-2.14

corrhi

-2.09

mohon

-1.91

 hundred

-1.89



-1.86

-1.82

 (!__

-1.76

-1.75

に入れる

-1.73

 Jugendliche

-1.69

POSITIVE LOGITS

は

2.47

がありますが

2.08

そして

2.00

というのも

1.98

もある

1.97

などが

1.93

を

1.92

1.84

も

1.81



1.81

Activations Density 0.078%