INDEX

Explanations

predicting equipment failure

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

 ningún

1.12

 nulla

1.10

 keiner

1.09

 nessuna

1.08

 nenhum

1.07

 žád

1.04

 plausible

1.04

 alcun

1.03

 completos

1.03

neither

1.01

POSITIVE LOGITS

Nowadays

1.46

Therefore

1.46

 Therefore

1.39

 lately

1.38

 Lately

1.37

 nowadays

1.36

 Nowadays

1.34

 그래서

1.29

 therefore

1.28

Meanwhile

1.26

Activations Density 0.101%