INDEX

Explanations

references to the letter 'L' in various contexts

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

 Petro

-0.07

hal

-0.06

iew

-0.06

Kid

-0.06

 Phot

-0.06

leigh

-0.06

oho

-0.06

Kid

-0.06

 residents

-0.05

 Micha

-0.05

POSITIVE LOGITS

azio

0.15

ucc

0.13

igure

0.12

arian

0.12

omb

0.11

amped

0.11

ario

0.11

ido

0.11

ucch

0.11

odi

0.10

Activations Density 0.005%