INDEX

Explanations

example phrases

The neuron strongly activates at the starts of parenthetical asides and at sentence boundaries (e.g. on an opening parenthesis or sentence-ending punctuation).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 परन्तु

1.00

 Однако

0.99

 然而

0.96

 However

0.95

 그러나

0.94

 परंतु

0.93

但

0.91

 لكن

0.90

 Cependant

0.87

 ولكن

0.85

POSITIVE LOGITS

 mesela

1.22

例えば

1.20

比如

1.18

比如说

1.16

たとえば

1.13

например

1.12

 misalnya

1.10

 например

1.05

比如說

1.05

 beispielsweise

1.04

Activations Density 0.291%