INDEX

Explanations

code block structure

The neuron strongly activates on hyphenated or multi-word compound adjectives and descriptive phrases (e.g. “hook-and-loop strap,” “highly professional team”).

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

↵↵↵↵↵↵↵↵↵↵↵

0.82

↵↵↵↵↵↵↵↵↵↵↵↵↵

0.80

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

0.79

↵↵↵↵↵↵↵

0.78

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

0.77

↵↵↵↵↵↵↵↵↵↵

0.76

↵↵↵↵↵↵↵↵↵

0.75

↵↵↵↵↵↵↵↵↵↵↵↵↵↵

0.73

↵↵↵↵↵↵↵↵

0.72

↵↵↵↵↵

0.71

POSITIVE LOGITS

IsIn

0.55

def

0.52

चीन

0.52

োজনে

0.52

deb

0.50

own

0.50

 frameborder

0.50

 blow

0.49

坚

0.49

className

0.49

Activations Density 0.046%