INDEX

Explanations

explaining what this is

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

2.97

lots

2.86

The

2.60

This

2.58

If

2.56

From

2.55

To

2.55

That

2.52

After

2.49

 That

2.47

POSITIVE LOGITS

ત્મ

3.35

⸙

3.23

 fateful

2.98

 মুহূর্তে

2.97

 misguided

2.96

 মুহুর্তে

2.91

{@

2.91

 momentous

2.87

féle

2.85

𝙧

2.85

Activations Density 1.355%