INDEX

Explanations

numeric sequences or references related to time or hierarchy

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

 sevent

-0.10

entieth

-0.09

 Seven

-0.09

XVI

-0.08

 Eight

-0.08

Six

-0.08

SIX

-0.08

åħ«

-0.08

 nineteen

-0.08

 nineteenth

-0.08

POSITIVE LOGITS

0.27

 twelve

0.21

 Twelve

0.20

Û±Û²

0.19

åįģäºĮ

0.19

XII

0.17

 dozen

0.16

Tw

0.15

tw

0.15

tw

0.14

Activations Density 0.035%