INDEX

Explanations

idiomatic phrases and outcomes

The neuron fires on abstract, nominalized words (i.e. nouns formed by suffixation, like “blame,” “stage,” “message,” “balance,” “story,” etc.).

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

denly

-1.20

 barva

-1.08

 trám

-1.02

 ständig

-1.01

 vám

-0.97

 sonriendo

-0.96

૬

-0.94

 venido

-0.92

 exitoso

-0.92

 siRNA

-0.92

POSITIVE LOGITS

of

3.38

 necessary

1.37

 that

1.20

 required

1.18

 statunit

1.16

 thereof

1.16

for

1.14

 needed

1.13

iest

1.06

 there

1.03

Activations Density 0.123%