INDEX

Explanations

Results

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ner

-0.96

iner

-0.82

INER

-0.82

NER

-0.82

SequentialGroup

-0.62

Cartney

-0.60

Results

-0.60

amine

-0.58

Bi

-0.52

fuge

-0.51

POSITIVE LOGITS

AndEndTag

0.81

[]:

0.71

 snippetHide

0.63

 defaultstate

0.63

Hochspringen

0.59

 engin

0.59

Clik

0.58

 Engine

0.56

 nahilalakip

0.56

 Incl

0.56

Activations Density 0.070%