INDEX

Explanations

where: phrase follows

This neuron fires on the tokens used in formula definitions—i.e. the “where … is …” phrases and their numeric or variable labels in mathematical expressions.

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

鈎

-0.93

葦

-0.92

تفسير

-0.90

熾

-0.90

ieš

-0.90

to

-0.89

Fär

-0.87

ΑΝ

-0.85

positorio

-0.84

 daha

-0.84

POSITIVE LOGITS

をして

0.97

ſelf

0.94

万

0.94

 möchtest

0.89

 adhesivo

0.88

atex

0.88

indrops

0.88

 الجز

0.87

 diſt

0.86

levé

0.86

Activations Density 0.018%