INDEX

Explanations

references to euphemisms and their implications

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Negative Logits

pun

-0.07

obo

-0.07

mans

-0.06

unit

-0.06

asks

-0.06

 tang

-0.05

mitt

-0.05

 interven

-0.05

 traditionally

-0.05

opic

-0.05

POSITIVE LOGITS

ayn

0.08

 misuse

0.08

 usage

0.08

 useClass

0.08

Î¬Ïĥ

0.07

nex

0.07

asel

0.07

_usage

0.07

Use

0.07

Usage

0.07

Activations Density 0.003%

No Known Activations