INDEX

Explanations

actions and their focus

The neuron activates on personal names—proper nouns referring to people.

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

おわりに

-1.59

 согласно

-1.38

 vriende

-1.35

 präsentieren

-1.33

 sostegno

-1.33

 niewiel

-1.30

tualmente

-1.27

žno

-1.24

 wezen

-1.23

 indruk

-1.20

POSITIVE LOGITS

and

1.51

we

1.36

if

1.34

so

1.30

 зимние

1.26

who

1.19

 potem

1.13

 հղ

1.11

 while

1.09

Activations Density 0.091%