INDEX

Explanations

references to individuals or groups in unfortunate circumstances

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

 Buchanan

-0.07

al

-0.07

ase

-0.06

erez

-0.06

å«Į

-0.06

alo

-0.06

 chod

-0.06

Ìĥ

-0.06

atis

-0.06

idable

-0.06

POSITIVE LOGITS

 souls

0.13

 soul

0.11

 thing

0.11

 Soul

0.09

sod

0.09

 Souls

0.09

Thing

0.08

 little

0.08

thing

0.08

 excuse

0.08

Activations Density 0.009%