INDEX

Explanations

state or action after subject

The neuron activates on capitalized named-entity tokens (e.g. people, organizations, places, or group names).

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

and

-1.73

or

-1.28

 which

-1.20

 kommunizieren

-1.12

 görüntüleri

-1.07

 ziekte

-1.06

 their

-1.05

ようで

-1.03

 that

-1.02

POSITIVE LOGITS

 also

1.40

 could

1.34

もあって

1.27

 still

1.21

 would

1.21

だろうか

1.19

 möjlighet

1.18

 recinto

1.16

iramente

1.15

 will

1.13

Activations Density 0.083%