INDEX

Explanations

No Explanations Found

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

 necessarily

1.72

𝐨

1.63

 "****

1.58

<unused2138>

1.56

<unused2130>

1.52

"<<

1.39

...";

1.38

...."

1.37

 ########

1.36

𝒐

1.36

POSITIVE LOGITS

And

1.44

và

1.43

1.35

And

1.31

AND

1.30

และ

1.29

irection

1.24

 และ

1.22

and

1.21

ssa

1.19

Activations Density 0.082%

No Known Activations