INDEX

Explanations

model acknowledging a request

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

اً

2.80

dua

2.42

만

2.38

داران

2.30

ر

2.27

ti

2.23

কার

2.20

cu

2.19

ا

2.17

キュリティ

2.16

POSITIVE LOGITS

ун

2.84

ocolate

2.72

ච

2.61

"।

2.58

ildren

2.57

大地

2.54

hhhh

2.49

loride

2.48

ायती

2.48

CharSequence

2.47

Activations Density 0.144%