INDEX

Explanations

rule of thumb

The neuron detects mentions of “rule of …” constructions, i.e. numerical or “rule of thumb” formulas.

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

 سنی

0.64

{}".

0.59

 exacta

0.57

лке

0.56

%";

0.56

wsp

0.54

?";

0.54

лко

0.54

 conclusively

0.54

TeV

0.54

POSITIVE LOGITS

 अनिल

0.57

sub

0.57

sum

0.57

ún

0.55

一下

0.54

getSub

0.54

角

0.53

以下

0.53

tzw

0.53

西北

0.52

Activations Density 0.000%