INDEX

Explanations

Match

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

Match

-0.82

 Match

-0.75

match

-0.68

 MATCH

-0.55

MATCH

-0.49

匹配

-0.48

 matching

-0.47

anz

-0.43

 match

-0.42

Matcher

-0.42

POSITIVE LOGITS

 Theſe

1.01

Efq

0.93

 Monfieur

0.92

 myſelf

0.92

 poffible

0.88

 poffe

0.88

ſelf

0.85

 uſed

0.84

 pleaſure

0.81

 themſelves

0.79

Activations Density 0.155%