INDEX

Explanations

people referring to self or others

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

 that

-1.79

-1.59

 trän

-1.24

lä

-1.16

пион

-1.15

adır

-1.13

 näm

-1.12

());

-1.10

нут

-1.09

เรา

-1.09

POSITIVE LOGITS

brtc

1.50

と思いますが

1.45

でしょうね

1.35

 also

1.33

 ähn

1.29

菈

1.28

 cazul

1.21

赑

1.21

也要

1.20

 แต่

1.17

Activations Density 0.051%