INDEX

Explanations

cultural exchange, designing, understanding

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

বিষয়

2.57

‌پ

2.41

ṓ

2.29

ผลิตภัณฑ์

2.28

ickým

2.24

বিষয়

2.23

álním

2.22

 سلاټ

2.18

řich

2.17

ผู้

2.14

POSITIVE LOGITS

 crap

2.81

 shit

2.66

ppl

2.65

 fucking

2.63

 fuck

2.61

 stuff

2.52

 crappy

2.52

 kinda

2.51

 fucked

2.50

 yada

2.45

Activations Density 2.619%