INDEX

Explanations

Physical interaction, Medical lines

references to TensorFlow (tf) in code and technical documentation.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

乆

-2.31

᠁

-2.31

𝗸

-2.16

-2.08

nologue

-2.06

рока

-2.03

ของ

-2.03



-1.99

畯

-1.96

 celebrar

-1.95

POSITIVE LOGITS

 comprehensively

2.88

2.84

as

2.59

2.56

 miraculously

2.52

www

2.44

 captivated

2.42

 suitably

2.39

 aquellos

2.38

 rigorously

2.34

Activations Density 0.002%