INDEX

Explanations

indifference and acceptance

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

ボリューム

-0.88

for

-0.79

と思い

-0.78

in

-0.77

感のある

-0.77

aces

-0.74

 then

-0.73

Hrs

-0.73

can

-0.71

ECONDS

-0.71

POSITIVE LOGITS

 calmly

1.44

 спокойно

1.27

 nonchal

1.25

 calm

1.24

 acceptance

1.15

 keinerlei

1.13

反而

1.12

 attitude

1.09

 shrug

1.06

むしろ

1.02

Activations Density 0.037%