INDEX

Explanations

developed by

The neuron detects formal disclaimer and warranty language (e.g. “provided,” “endorsement,” “products,” “services”) typical of legal/terms-of-use boilerplate.

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

នៃ

0.53

immt

0.52

лә

0.51

രെ

0.49

әп

0.49

волю

0.47

 বিপর

0.47

лык

0.47

さに

0.47

ියේ

0.46

POSITIVE LOGITS

 oleh

4.47

by

4.16

 توسط

3.91

โดย

3.25

by

3.11

 tarafından

3.00

 द्वारा

2.84

 بواسطة

2.72

によって

2.66

 দ্বারা

2.61

Activations Density 1.321%