INDEX

Explanations

quantified completed actions

The neuron is detecting numeric quantity tokens (i.e., measurements or counts).

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

蹚

-1.41

 creativos

-1.32

 koruyucu

-1.29

 tenho

-1.28

 burbujas

-1.27

useEffect

-1.27

</u>

-1.27

粜

-1.25

 koko

-1.24

 innebär

-1.23

POSITIVE LOGITS

in

2.95

 from

2.34

 during

1.98

by

1.92

at

1.76

 between

1.74

 with

1.67

for

1.60

on

1.59

 through

1.45

Activations Density 0.029%