INDEX

Explanations

being or existence

The neuron fires most strongly on unusually rare or low‐frequency words.

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

なんですよ

-1.00

elezo

-0.92

なんですね

-0.87

рующие

-0.86

 olmaz

-0.86

 kellett

-0.86

 副本

-0.85

assertRaises

-0.84

timme

-0.84

fetera

-0.84

POSITIVE LOGITS

is

4.19

 είναι

2.64

was

2.53

are

2.52

 है

2.03

是

1.84

 છે

1.83

的是

1.80

 ਹੈ

1.77

be

1.63

Activations Density 0.005%