INDEX

Explanations

anticipate or plan for future

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 suspicious

-0.10

confirmation

-0.10

 curtains

-0.10

-summary

-0.10

 suspicion

-0.10

 confidential

-0.10

 Summary

-0.10

(CC

-0.10

 confirmed

-0.10

 confirmation

-0.10

POSITIVE LOGITS

 anticipated

0.34

 fores

0.34

 unfore

0.33

 foresee

0.32

 anticipate

0.32

 anticipating

0.30

anticipated

0.28

 anticip

0.26

 contingency

0.26

 foreseeable

0.24

Activations Density 0.425%