INDEX

Explanations

разработка, размеры, разбивка

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

̓

2.22

isham

2.20

 speechless

2.19

íj

2.17

tics

2.15

 richiesta

2.15

estä

2.13

جر

2.12

 glaciers

2.12

langs

2.11

POSITIVE LOGITS

Ошибка

1.23

詳細は

1.12

ORITY

1.11

ЕД

1.10

================

1.07

****************

1.05

 MUCH

1.03

PLAY

1.03

 unsafe

1.02

 mnog

0.99

Activations Density 0.042%