INDEX

Explanations

links to descriptions

The neuron detects parts of web links or URLs (e.g., tokens from “http://…”, domain names, and path segments).

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

-2.44

-1.98

It

-1.91

-1.84

If

-1.77

-1.75

-1.72

we

-1.72

 трябва

-1.68

 Hinblick

-1.66

POSITIVE LOGITS

the

1.77

ляции

1.62

Beskrivelse

1.61

ieke

1.59

of

1.57

Transk

1.54

 inaweza

1.51

 ممکن

1.49

ότη

1.47

謬

1.47

Activations Density 0.017%