INDEX

Explanations

studies and services

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 estimate

-1.25

 Estimates

-1.04

 estimates

-1.04

 Estimate

-1.01

Estimate

-0.93

setVerticalGroup

-0.92

 estimation

-0.85

Estimates

-0.84

NUMX

-0.82

estimate

-0.81

POSITIVE LOGITS

0.55

 célèbres

0.50

 giusti

0.47

0.46

 paix

0.45

 soutenu

0.45

 macchine

0.45

 braccia

0.45

 suivants

0.44

 connues

0.44

Activations Density 0.007%