INDEX

Explanations

causes or helps to make

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

 where

-1.36

 before

-1.33

 while

-1.24

 using

-1.20

 with

-1.16

 However

-1.13

on

-1.08

as

-1.06

 after

-1.06

 without

-1.05

POSITIVE LOGITS

 immense

1.21

 Provides

1.13

慢慢

1.13

 colossal

1.12

 hugely

1.11

 significantly

1.10

 also

1.09

 broader

1.07

 tremendous

1.05

 quirky

1.04

Activations Density 0.035%