INDEX

Explanations

article followed by noun

The neuron fires strongly on multiword idiomatic or phrasal expressions (e.g. “dealt the short end,” “steppin’ out of,” “look under the hood,” “composing the…”) rather than isolated 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

of

-1.46

 рекомендации

-1.38

tuto

-1.35

 verhind

-1.33

 EIGHT

-1.33

 советы

-1.25

 revamped

-1.23

 juſ

-1.20

to

-1.18

-1.16

POSITIVE LOGITS

 this

1.95

 große

1.63

 some

1.57

 many

1.41

one

1.39

 There

1.38

 sizable

1.38

 очень

1.37

 there

1.35

 Usually

1.35

Activations Density 0.028%