INDEX

Explanations

mentions of behind-the-scenes content

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.Îł

-0.06

oire

-0.06

SKI

-0.06

 //!<

-0.06

_raise

-0.05

 variation

-0.05

abwe

-0.05

 Harrison

-0.05

/umd

-0.05

Ð½Ð¾ÑģÑı

-0.05

POSITIVE LOGITS

-scenes

0.16

 scenes

0.11

scenes

0.11

 scene

0.10

åľº

0.09

 Scenes

0.09

-camera

0.09

scene

0.09

enes

0.08

åł´

0.08

Activations Density 0.003%