INDEX

Explanations

all of which

The neuron activates on general quantifier words (like “all,” “many,” “some,” etc.) that introduce sweeping statements about quantity.

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

ཿ

-1.61

 aguj

-1.52

鵙

-1.45

 salonu

-1.44

𝐾

-1.41

cím

-1.41

 botellas

-1.38

絝

-1.38

 koncep

-1.38

 ahorrar

-1.38

POSITIVE LOGITS

of

5.84

 with

3.27

 which

2.48

 from

2.17

for

2.16

 without

1.95

in

1.93

at

1.88

 having

1.59

on

1.57

Activations Density 0.054%