INDEX

Explanations

neural network flatten layer

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

ी

2.14

رل

2.06

"،

2.03

!".

2.01

kipun

1.96

"!

1.95

 '\\;'

1.84

 throwIf

1.79

"`

1.79

drain

1.76

POSITIVE LOGITS

 kuten

2.04

 jonka

1.77

↵

1.63

 illetve

1.57

 joka

1.54

 czyli

1.53

 jossa

1.52

teurs

1.47

 että

1.46

</em>

1.46

Activations Density 0.023%