INDEX

Explanations

code explanations and examples

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

 amts

1.15

Église

1.04

 fetal

1.03

osil

0.98

новый

0.97

 Iglesia

0.94

 hautes

0.93

 Tochter

0.92

гар

0.91

 رجسٹر

0.91

POSITIVE LOGITS

 Explanation

2.01

Explanation

1.99

 السؤال

1.92

explanation

1.92

样例

1.86

思路

1.80

 explanation

1.79

 SOLUTION

1.77

解释

1.71

 Solution

1.69

Activations Density 0.031%