INDEX

Explanations

code and technical descriptions

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

整数

1.29

मराह

1.27

 wajah

1.19

ग्रियों

1.18

раху

1.17

 istedi

1.15

روف

1.15

omeg

1.15

 Fragen

1.13

 frü

1.13

POSITIVE LOGITS

 down

1.55

down

1.35

ลง

1.23

Down

1.21

downs

1.19

eree

1.16

 Down

1.08

ing

1.05

 DOWN

1.04

ermost

1.02

Activations Density 0.617%