INDEX

Explanations

easy and simple descriptions

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

 Favor

-0.10

fault

-0.10

assin

-0.10

 phenomena

-0.09

uld

-0.09

ula

-0.09

favor

-0.09

 stunt

-0.09

 Fault

-0.09

andan

-0.09

POSITIVE LOGITS

 easy

0.17

 Easy

0.15

_easy

0.15

easy

0.14

Easy

0.13

 simple

0.12

 Simple

0.12

-simple

0.11

 easier

0.11

/simple

0.11

Activations Density 0.018%