INDEX

Explanations

use of

The neuron strongly fires on the noun “use” (and its close variants) whenever the text is talking about the use of something.

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

рабо

0.62

த்திலிருந்து

0.61

ással

0.59

 Побе

0.59

ರುವುದ

0.59

變化

0.59

線を

0.58

ધું

0.58

角落

0.58

 Счет

0.57

POSITIVE LOGITS

of

1.59

 của

1.17

 của

1.09

 ofthe

1.06

ของ

1.02

 של

1.02

of

0.92

 των

0.81

ຂອງ

0.81

pf

0.79

Activations Density 0.004%