INDEX

Explanations

isolate, provide, modify, training

The neuron strongly fires on numeric figures and statistical values (i.e. tokens carrying digits or numbers).

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

or

0.81

ট

0.79

 Sheila

0.77

0.74

大多

0.74

 inactiv

0.73

Myl

0.73

 речь

0.72

 Celia

0.72

 Dorian

0.71

POSITIVE LOGITS

liers

0.72

 respiratoires

0.72

ristmas

0.71

ђу

0.70

 summed

0.68

 rusty

0.68

ansion

0.67

mainan

0.67

asjoner

0.67

 newest

0.66

Activations Density 0.000%