INDEX

Explanations

given data and information

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

 такая

1.93

 употре

1.82

 aprovech

1.72

أ

1.66

 останавли

1.64

 ktoś

1.62

ذ

1.62

由于

1.60

 это

1.60

 zadr

1.58

POSITIVE LOGITS

 timeframe

2.41

 ‘’

2.33

 trio

2.32

ussians

2.31

 newfound

2.26

 ISNI

2.23

 timeline

2.22

duo

2.22

 "));

2.21

 whopping

2.18

Activations Density 0.217%