INDEX

Explanations

named code identifier

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 році

2.77

йын

2.73

angkah

2.71

ați

2.57

+'-

2.55

šao

2.55

нению

2.54

จัย

2.52

rometer

2.51

 noastră

2.49

POSITIVE LOGITS

𝐧

3.66

𝐫

3.43

્ય

3.28

𝐞

3.24

𝐨

3.22

𝐚

3.13

︎

3.12

teen

3.09

𝐭

3.00

theless

2.97

Activations Density 0.530%