INDEX

Explanations

now

The neuron strongly activates on the word “now,” flagging that temporal/discourse marker wherever it appears.

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

↵

-1.70

↵↵

-1.39

 today

-1.31

-1.27

of

-1.27

but

-1.24

 actualmente

-1.22

↵↵↵

-1.16

については

-1.13

POSITIVE LOGITS

the

1.70

titial

1.64

ꔹ

1.60

 vlasy

1.54

🤦

1.52

 mostrou

1.51

ية

1.50

🥲

1.49

埚

1.48

銠

1.48

Activations Density 0.020%