INDEX

Explanations

experimental research methods

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

1.39

 एन

1.29

有效期

1.25

ƌ

1.22

エ

1.20

Rl

1.20

ary

1.19

 रमेश

1.18

1.17

1.16

POSITIVE LOGITS

 gluing

1.54

 muzzle

1.46

 underlying

1.46

<unused298>

1.46

 geom

1.39

 edgy

1.35

ိပ်

1.34

 facere

1.32

സ്യ

1.32

<unused1839>

1.30

Activations Density 0.065%