INDEX

Explanations

showing or producing results

The neuron responds to the main action verbs describing what the user or code is doing (e.g., “showing,” “written,” “sending,” “creating,” “got,” etc.).

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

my

-1.44



-1.33

"")

-1.33

 usamos

-1.30

 nasz

-1.27

cembre

-1.24

 réjou

-1.22

 hés

-1.19

Ғ

-1.19

 jego

-1.16

POSITIVE LOGITS

 such

2.47

 different

2.14

 also

1.91

 various

1.63

 very

1.59

 similar

1.42

 tikai

1.41

 такі

1.41

not

1.41

Activations Density 0.172%