INDEX

Explanations

mentions or references related to findings, suggestions, or inferences drawn from data, results, or observations

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 abzu

-0.77

alamu

-0.73

 якості

-0.68

MultipartFile

-0.68

 zamiast

-0.66

Zalety

-0.66

 multipurpose

-0.62

 Multi

-0.62

 отношению

-0.61

 lunghi

-0.61

POSITIVE LOGITS

 Sugges

1.27

 claims

1.23

 claimed

1.21

 claim

1.20

 believed

1.18

 suggest

1.16

 suggested

1.16

 SUGGEST

1.16

 CLAIM

1.16

 Claims

1.13

Activations Density 9.587%