INDEX

Explanations

cram, Cramer, handleChange, iomanip, barometer

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 untuk

-3.20

 để

-2.72

 ignite

-2.53

-2.48

vantaged

-2.42

низ

-2.39

 maximize

-2.38

↵↵

-2.34

 Verhält

-2.33

POSITIVE LOGITS

一颗

2.77

丝毫

2.38

がなく

2.36

がたくさん

2.30

这事

2.28



2.27

簠

2.25

整个人

2.23

榇

2.19

 organes

2.13

Activations Density 0.010%