INDEX

Explanations

Therefore, data, queries, SELECT

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 this

-1.30

といけない

-1.09

ialog

-1.08

 tohoto

-1.05

 другое

-1.05

жима

-1.04

 dlatego

-1.03

роятно

-1.03

iramente

-1.03

場合は

-1.03

POSITIVE LOGITS

 each

2.48

any

2.25

if

2.22

Each

2.13

Any

1.90

 there

1.89

 only

1.80

 every

1.80

每个

1.73

 when

1.70

Activations Density 0.208%