INDEX

Explanations

instances of the word "fine" or variations of it, indicating satisfaction or approval 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

anca

-0.08

rez

-0.07

rey

-0.07

dens

-0.07

aurus

-0.06

dale

-0.06

redo

-0.06

rence

-0.06

wo

-0.06

rei

-0.06

POSITIVE LOGITS

akan

0.08

forth

0.08

acre

0.07

/error

0.07

-sized

0.07

ØªØ±ÛĮ

0.07

leck

0.07

lahoma

0.07

egal

0.07

hosts

0.07

Activations Density 0.004%