INDEX

Explanations

per

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

per

-1.14

Per

-0.73

Per

-0.63

-0.62

in

-0.61

PER

-0.59

for

-0.54

so

-0.54

up

-0.54

and

-0.53

POSITIVE LOGITS

 pleaſure

0.79

 purpoſe

0.78

 fevere

0.77

 myſelf

0.77

 perfons

0.77

 Majefty

0.75

 poffible

0.74

 ſmall

0.73

 poffe

0.73

 Chriftian

0.72

Activations Density 0.032%