INDEX

Explanations

No Explanations Found

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

 blessés

-1.16

曲を

-1.15

曲は

-1.09

 Februari

-1.09

ATRE

-1.06

 générales

-1.04

rrrrrrrr

-1.03

🧡

-1.02

 registre

-1.02

💔

-1.01

POSITIVE LOGITS

and

1.48

 modestly

1.16

 magically

1.16

 regionally

1.13

 albeit

1.13

 intensified

1.11

 tougher

1.09

 admittedly

1.09

ところに

1.08

 painstaking

1.07

Activations Density 0.000%

No Known Activations

This feature has no known activations.