INDEX

Explanations

phone numbers

The neuron fires on multi‐digit numeric tokens (e.g. parts of phone numbers or other standalone numbers).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 évé

-1.51

 balón

-1.43

 norvég

-1.41

 salón

-1.41

-1.36

 kabát

-1.27

 márm

-1.23

औ

-1.23

 competir

-1.23

 their

-1.22

POSITIVE LOGITS

見ると

1.13

We

1.05

 Именно

1.03

ip

1.02

‐

1.01

もちろん

1.00

 чаще

1.00

 некоторых

0.99

 испыты

0.98

Activations Density 0.001%