INDEX

Explanations

from 'react' or meta charset

The neuron fires on multi‐digit numeric tokens (especially decimals), such as the numbers and DOIs in scientific references.

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

sab

0.54

 kulland

0.49

哈哈

0.47

hau

0.46

 memnun

0.46

 kısa

0.46

荣誉

0.45

 yapmış

0.44

 baina

0.44

綺麗な

0.44

POSITIVE LOGITS

vigor

0.49

 Cloudy

0.45

 SQLException

0.45

 Hinton

0.45

 Highway

0.44

Highway

0.44

ﻔ

0.44

NEAR

0.44

excitation

0.43

 mín

0.42

Activations Density 0.000%