INDEX

Explanations

onto physical objects

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bury

-0.09

 barg

-0.09

 locally

-0.08

 cheek

-0.08

 ponder

-0.08

cel

-0.08

ovan

-0.08

 Ships

-0.08

 hull

-0.07

 drifting

-0.07

POSITIVE LOGITS

 spike

0.13

edList

0.12

 sharpen

0.12

 spiked

0.12

 spikes

0.11

Ost

0.11

 shaft

0.11

 mounted

0.10

avour

0.10

mounted

0.10

Activations Density 0.019%