INDEX

Explanations

characters and symbols used in various encoding formats

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

WTF

-0.09

 cazzo

-0.08

 fuck

-0.08

 bullshit

-0.07

NYC

-0.07

ä¿º

-0.07

Fuck

-0.07

fuck

-0.07

 fucking

-0.07

 fucked

-0.07

POSITIVE LOGITS

Mir

0.10

Fel

0.09

Mir

0.08

Hil

0.08

 Miss

0.08

 Fifth

0.07

 sixth

0.06

Miss

0.06

 mirrors

0.06

Activations Density 0.004%