INDEX

Explanations

code and text fragments

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

-0.21

hibited

-0.20

ocrats

-0.20

èĩ´ãģĹãģ¾ãģĻ

-0.20

iente

-0.19

onent

-0.19

 Murphy

-0.19

åĪ«

-0.18

Erf

-0.18

POSITIVE LOGITS

 Decompiled

0.23

 Interracial

0.23

**/↵↵

0.21

fkk

0.21

 WHETHER

0.20

arendra

0.20

:"#

0.20

.moveToNext

0.20

æĦĽæĥħ

0.20

ourd

0.20

Activations Density 0.002%