INDEX

Explanations

particularly followed by descriptive adjectives

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

 proch

2.32

 zwar

2.23

誨

2.14

 stimmen

2.12

koj

2.11

 houd

2.10

ന്‍

2.09

 pouls

2.07

 tyle

2.06

ਾਰ

2.05

POSITIVE LOGITS

kent

2.21

remely

2.19

л

2.03

лта

1.88

лната

1.72

ی

1.72

ureux

1.71

至

1.71

్రియ

1.70

kings

1.69

Activations Density 0.016%