INDEX

Explanations

reaction

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ahu

-0.14

 Spoon

-0.11

 unrecognized

-0.11

reminder

-0.11

ATT

-0.11

Suit

-0.11

mue

-0.11

oS

-0.11

 Forgotten

-0.11

ATT

-0.11

POSITIVE LOGITS

 reaction

0.22

 Shock

0.20

 Reaction

0.19

 shock

0.18

 reactions

0.18

 shocked

0.17

Reaction

0.17

reaction

0.16

Shock

0.16

 reacted

0.16

Activations Density 0.031%