INDEX

Explanations

comparison words

The neuron responds to comparative and superlative adjectives (like “bigger,” “smaller,” “largest,” and “large”) used in numeric or quantitative comparisons.

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

 Conſ

-2.33

 Inſ

-2.22

瞜

-2.03

狆

-1.96

迸

-1.93

ophones

-1.91

 Diſ

-1.89

 Reſ

-1.88

 火鍋

-1.87



-1.87

POSITIVE LOGITS

 what

2.19

the

2.11

 from

2.05

as

2.02

in

1.98

 rapidly

1.98

 whose

1.97

 with

1.97

 first

1.97

 that

1.94

Activations Density 0.039%