INDEX

Explanations

specific word pairings

explaining specific meanings or references

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

direccion

0.50

bouw

0.49

 않는

0.48

 gasp

0.48

 Gunners

0.47

 succumb

0.46

 fissures

0.46

 Николай

0.45

 بیٹھ

0.45

 watery

0.45

POSITIVE LOGITS

UL

0.54

etí

0.45

 Nearest

0.45

MAN

0.45

ेंड

0.45

ВА

0.45

ালো

0.44

 Sedan

0.43

 Distributions

0.43

idelberg

0.42

Activations Density 0.001%