INDEX

Explanations

web links/code

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 usage

-1.16

 Usage

-0.88

 usages

-0.83

av

-0.82

Usage

-0.72

usage

-0.72

 utilization

-0.71

use

-0.69

 USAGE

-0.66

 drag

-0.63

POSITIVE LOGITS

0.66

 quæ

0.65

 whoſe

0.60

 againſt

0.58

 Fiche

0.56

 DriverManager

0.54

 nôtre

0.53

 prouver

0.53

 verrez

0.52

 juſt

0.52

Activations Density 0.051%