INDEX

Explanations

/ followed by numbers

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cter

1.25

 ίδ

1.19

gest

1.13

 azon

1.12

估计

1.08

 allem

1.08

设计

1.06

adjoint

1.05

gem

1.05

 vận

1.04

POSITIVE LOGITS

ęb

1.57

 arre

1.27

持って

1.27

 bahas

1.27

蔑

1.23

喀

1.20

茹

1.18

そこに

1.15

ába

1.14

 Blot

1.14

Activations Density 0.094%