INDEX

Explanations

ke

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ke

-0.68

itent

-0.65

oya

-0.53

ी

-0.49

capital

-0.49

hip

-0.47

 متعلقه

-0.47

penn

-0.47

ècie

-0.46

hips

-0.46

POSITIVE LOGITS

<bos>

0.63

TagMode

0.62

 Reverso

0.62

帖最后由

0.58

 defaultstate

0.58

(!__

0.57

apimachinery

0.56

Jegyzetek

0.56

>{@

0.54

 EconPapers

0.54

Activations Density 0.028%

No Known Activations