INDEX

Explanations

broke

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sust

-0.30

::{

-0.27

 taps

-0.26

icas

-0.25

!!

-0.25

Met

-0.25

Is

-0.24

 Singleton

-0.24

!!

-0.24

ISC

-0.24

POSITIVE LOGITS

åĪĻ

0.69

å°±å¿ħé¡»

0.57

å°±è¦ģ

0.55

æĪĳä»¬å°±

0.54

å°±è¯´

0.54

å°±æĬĬ

0.52

å°±å¾Ĺ

0.51

å°±è¢«

0.51

å°±

0.50

å°±æ²¡æľī

0.49

Activations Density 0.003%