INDEX

Explanations

making legal claims and arguments

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 directive

-0.09

ogo

-0.09

ucken

-0.09

 administering

-0.09

 governing

-0.09

onia

-0.08

/dialog

-0.08

 Decide

-0.08

 deciding

-0.08

 fixing

-0.08

POSITIVE LOGITS

 claim

0.14

 claiming

0.13

claim

0.12

 unsuccessfully

0.12

 claims

0.12

 unsupported

0.11

unsupported

0.11

 Claim

0.11

 argue

0.11

 claimed

0.10

Activations Density 0.770%