INDEX

Explanations

hydrophobic

The neuron detects the term "hydrophobic" and related chemical/biochemical descriptions of molecules with water-repelling properties.

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

轳

-2.84

蓠

-2.23

鈰

-2.05

which

-1.98

 selben

-1.95

踭

-1.91

 Ensuring

-1.90

.’

-1.88

>\<

-1.84

 dergleichen

-1.83

POSITIVE LOGITS

2.06

 العاده

2.02

鷓

2.00

 mogen

1.90

了一

1.90

1.88

einf

1.85

 eloquent

1.84

Seven

1.77

Activations Density 0.016%