INDEX

Explanations

specific numeric or coded information and key indicators of performance metrics

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

 Clients

-0.07

 Hyde

-0.07

 clients

-0.07

marketing

-0.07

 marketing

-0.07

 client

-0.07

Marketing

-0.07

 CLIENT

-0.07

 Marketing

-0.06

clients

-0.06

POSITIVE LOGITS

 Cancer

0.10

 transl

0.09

 data

0.09

 datasets

0.09

 cancer

0.08

 computational

0.08

 dataset

0.08

 Data

0.08

 klin

0.07

 researchers

0.07

Activations Density 0.003%