INDEX

Explanations

series by Brandon Sanderson

The neuron specializes in spotting standalone numeric tokens (e.g. dates, times, addresses, counts or statistics).

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

 sejumlah

0.79

 instituições

0.74

 beragam

0.71

 thiab

0.71

umsal

0.70

 разнообраз

0.69

ስቃሴ

0.67

 отдельных

0.66

 działa

0.66

だけでなく

0.65

POSITIVE LOGITS

which

1.30

 which

1.29

 который

1.20

 которая

1.09

 WHICH

1.08

 която

1.01

 lequel

1.00

 welches

0.98

 котором

0.96

 який

0.96

Activations Density 0.345%