INDEX

Explanations

cookware and cooking

The neuron primarily picks up on numbers and numeric tokens (e.g. standalone digits, numeric units, measurements).

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

 CONCRETE

-0.84

 Cement

-0.83

 rotting

-0.77

 cement

-0.76

ditos

-0.75

Lleg

-0.73

自家製

-0.73

ヤマハ

-0.73

 Excise

-0.72

 бампер

-0.72

POSITIVE LOGITS

 cookware

1.44

 nons

1.24

 coating

1.18

nons

1.13

 cooking

1.09

coating

1.07

Induction

1.02

pan

1.02

 PTFE

1.00

 pans

0.96

Activations Density 0.005%