INDEX

Explanations

too much or too little

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

اء

3.17

ยนต์

2.98

𝐞

2.81

𝐚

2.60

st

2.56

č

2.46

ık

2.44

다고

2.41

𝐮

2.37

the

2.32

POSITIVE LOGITS

া

2.40

 Kudos

2.34

pportun

2.33

𝓁

2.27

𝒹

2.25

ристи

2.24

HandleFunc

2.22

ഴിഞ്ഞ

2.22

 Thames

2.22

 brach

2.19

Activations Density 0.157%