INDEX

Explanations

weight and characteristics

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

eki

2.56

iotics

2.54

𝔰

2.52

<bos>

2.50

ためには

2.49

hans

2.40

ripción

2.32

 inductor

2.30

equ

2.25

 pictures

2.25

POSITIVE LOGITS

嗦

2.66

Ia

2.61

arters

2.59

nuxtjs

2.51

썸

2.50

▟

2.46

ביר

2.44

UseDebug

2.43

صه

2.37

μένο

2.35

Activations Density 0.563%