INDEX

Explanations

Earth's atmosphere

The neuron activates on mentions of “atmosphere” (and closely related scientific/atmospheric context terms).

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

这些

-1.23

ARROLLO

-1.22

LITERATURE

-1.18

 boxeo

-1.17

 jadwal

-1.16

),

-1.15

UNIDENTIFIED

-1.15

ărilor

-1.11

prende

-1.11

тории

-1.10

POSITIVE LOGITS

 FontWeight

1.31

it

1.27

1.25

Sad

1.20

and

1.18

 RESPOND

1.14

not

1.13

 Another

1.13

inqui

1.13

att

1.11

Activations Density 0.064%