INDEX

Explanations

The neuron strongly activates on vivid descriptive words—especially sensory and qualitative adjectives (and related descriptors) that characterize taste, smell, texture, size, color, or other salient properties.

adjectives describing qualities

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

 boughs

-1.00

 لدي

-0.96

 conmigo

-0.95

 qualunque

-0.92

unno

-0.90

 tambin

-0.89

更高

-0.88

 Geheimnis

-0.86

 correspondingly

-0.85

 further

-0.85

POSITIVE LOGITS

ittää

1.27

itzenden

1.13

 ferie

1.08

 sodio

1.06

 tafel

1.05

čnost

1.05

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.04

emde

1.03

 tirs

1.03

eeu

1.01

Activations Density 0.066%