INDEX

Explanations

finalist

The neuron activates on mentions of “finalist” or “finalists,” i.e. references to competitors who have reached the final stage.

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

we

-2.63

 einigen

-2.44

Britannique

-2.31

Our

-2.17

 Isten

-2.14

锱

-2.13

ارتمان

-2.13

ydı

-2.09

躡

-2.06

 Tetapi

-2.03

POSITIVE LOGITS

3.33

2.91

是很

2.69

Not

2.53

Of

2.42

2.41

All

2.28

No

2.23

ですが

2.20

Activations Density 0.004%