INDEX

Explanations

family relationships and code

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

An

1.80

Se

1.75

ف

1.72

SE

1.72

Te

1.71

гії

1.71

vgl

1.66

สัญญาณ

1.64

お

1.62

The

1.59

POSITIVE LOGITS

woe

2.41

ac

2.19

 tripped

2.17

 struggled

2.12

 stricken

2.10

 knocked

2.10

 binaries

2.06

 stepped

2.01

 capacitor

1.97

ູ້

1.94

Activations Density 0.000%