INDEX

Explanations

visibility

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

 visible

-0.81

Visible

-0.72

 possible

-0.71

visible

-0.70

 posibles

-0.68

possible

-0.66

ing

-0.64

 possíveis

-0.61

 possível

-0.60

 möjligt

-0.60

POSITIVE LOGITS

 Theſe

0.99

Efq

0.96

 myſelf

0.87

 Anſ

0.81

 Jefus

0.81

 themſelves

0.78

 againſt

0.77

 Beſ

0.77

 pleaſure

0.77

 Monfieur

0.77

Activations Density 0.083%