INDEX

Explanations

figure references

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

水墨

-0.82

Beschrijving

-0.79

 engineer

-0.78

规模

-0.78

vaná

-0.78

Begriff

-0.77

濱

-0.77

Nicely

-0.76

出品

-0.76

 Conception

-0.76

POSITIVE LOGITS

 hundred

1.12

 hundert

1.06

ten

0.98

 dzies

0.91

 ২০

0.90

hundred

0.84

 деся

0.81

 Interference

0.81

 десять

0.80

۰۱

0.80

Activations Density 0.089%