INDEX

Explanations

path to

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

 продолжа

0.86

praw

0.83

**********

0.82

と呼ばれる

0.82

హా

0.81

FALSE

0.81

idium

0.81

islava

0.81

*********

0.80

 वैष्ण

0.80

POSITIVE LOGITS

Oh

1.08

 Compared

1.07

 Lookup

1.04

 Look

1.04

 externalities

1.00

 Here

0.99

 lookup

0.98

Now

0.97

Wow

0.96

 Based

0.94

Activations Density 0.077%