INDEX

Explanations

code and technical terms

This neuron responds to sequences of hexadecimal-like characters and numbers (e.g. cache keys, memory or identifier strings).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

鉚

0.53

ध्ये

0.52

plastics

0.52

agricultural

0.50

illustrated

0.50

िलं

0.50

PERATURE

0.49

的有

0.49

प्टेंबर

0.49

 कार्यकार

0.49

POSITIVE LOGITS

 BufferedReader

0.71

//

0.67

 Setiap

0.66

<0x0D>

0.65

if

0.63

 Bağ

0.61

 wannan

0.59

 ardından

0.58

 UpdateWindow

0.58

BufferedReader

0.57

Activations Density 0.002%