INDEX

Explanations

terms related to unauthorized usage and copyright restrictions

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

ikel

-0.06

Ù¾Ø±

-0.06

splash

-0.06

 Contracts

-0.06

wich

-0.06

onte

-0.06

 undert

-0.06

 Addresses

-0.06

 Ø§ÙĦØ£ÙħØ±

-0.06

POSITIVE LOGITS

 reproduce

0.08

 reproduction

0.07

allas

0.07

 reprodu

0.07

 Linh

0.07

use

0.07

inh

0.07

IFE

0.07

 attempt

0.07

ilib

0.07

Activations Density 0.005%