INDEX

Explanations

at least

The neuron specifically fires on the phrase “at least.”

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

 même

-1.69

 également

-1.56

因为

-1.54

其實

-1.50

actually

-1.49

 навіть

-1.48

 calcetines

-1.43

因為

-1.41

 även

-1.41

 גם

-1.40

POSITIVE LOGITS

for

2.17

 until

2.08

in

1.83

 when

1.82

 from

1.59

on

1.57

if

1.47

 according

1.45

to

1.41

 during

1.41

Activations Density 0.018%