INDEX

Explanations

but, and, then

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

😈

-1.78

 that

-1.77

vantaged

-1.53

 AMERIC

-1.51

after

-1.49

 after

-1.48

τως

-1.44

 MICHIGAN

-1.44

 montagem

-1.34

if

-1.33

POSITIVE LOGITS

</td>

1.63

Additionally

1.57

Moreover

1.53

Regarding

1.46

 unending

1.45

Similarly

1.41

Furthermore

1.40

 این

1.39

}(

1.37

٧

1.35

Activations Density 0.029%