INDEX

Explanations

rather than

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

cog

-0.10

pov

-0.10

inel

-0.09

 Hayes

-0.09

inde

-0.09

 Wolff

-0.09

 moreover

-0.09

 estoy

-0.09

 woll

-0.08

elay

-0.08

POSITIVE LOGITS

 risk

0.15

 risking

0.15

 trying

0.13

risk

0.13

 worrying

0.13

 having

0.12

 attempting

0.12

try

0.12

 attempt

0.12

 bother

0.12

Activations Density 0.030%