INDEX

Explanations

court of appeals

The neuron fires on long runs or blocks of whitespace (large sequences of spaces or indentation).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

tembre

-1.23

<bos>

-1.17

to

-1.14

を用

-1.13

 others

-1.13

lorin

-1.09

oprene

-1.02

flikt

-1.01

 here

-0.99

its

-0.98

POSITIVE LOGITS

ADORES

1.25

SBATCH

1.18

THE

1.16

.”

1.16

 africains

1.16

 soñar

1.15

excellents

1.14

.’”

1.10

gië

1.10

1.09

Activations Density 0.002%