INDEX

Explanations

too + adjective/adverb

The neuron consistently fires on the word “too,” identifying instances of that intensifier/adverb.

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

re

-2.16

able

-1.93

for

-1.79

-1.76

of

-1.73

and

-1.71

又は

-1.70

return

-1.66

-1.65

-1.63

POSITIVE LOGITS

嵓

2.16

푠

1.98

ܔ

1.82

 relativi

1.81

grandfather

1.73

ᆞ

1.72

鹀

1.69

どうやら

1.68

accident

1.67

 parachoque

1.67

Activations Density 0.008%