INDEX

Explanations

key names and the sequences of their actions and relationships throughout their lives

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

á»©

-0.08

reuse

-0.08

Ãłm

-0.07

sap

-0.07

Î¼ÎŃ

-0.07

recover

-0.07

RectTransform

-0.07

ledo

-0.07

enberg

-0.07

Å¯st

-0.07

POSITIVE LOGITS

 move

0.11

move

0.10

 moved

0.10

 moves

0.09

 Move

0.09

Move

0.09

 moving

0.08

-move

0.08

 cities

0.08

Moving

0.08

Activations Density 0.026%