INDEX

Explanations

identifying relevant content

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ﻲ

2.98

紝

2.61

্বল

2.60

்ட்

2.51

ี

2.43

ারা

2.42

ﺔ

2.40

 wavelet

2.28

Endian

2.24

ഹ്ലാദ

2.20

POSITIVE LOGITS

<bos>

4.04

geon

2.36

opia

2.34

cky

2.33

썩

2.24

passage

2.21

जरीवाल

2.11

2.08

pl

2.07

eous

2.05

Activations Density 0.074%