INDEX

Explanations

DTD XHTML

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

reten

-0.79

Duf

-0.74

 Housekeeping

-0.71

 pourtant

-0.69

ándote

-0.68

jir

-0.66

AGON

-0.66

 Aspect

-0.66

 morti

-0.65

 Vulnerability

-0.65

POSITIVE LOGITS



0.68

়

0.63

quant

0.62

 kapcsolat

0.61

Pa

0.61

 galt

0.61

мал

0.60

\%,

0.60

 глоба

0.60

 whenever

0.59

Activations Density 0.082%