INDEX

Explanations

first in a series of

This neuron detects numeric quantifiers and ordinal positioning phrases (e.g. “first of four,” “first of three,” etc.) indicating an item’s place in a series.

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

 lately

-1.24

 trecut

-1.16

 already

-1.13

 последнее

-1.12

 τελευτα

-1.10

 acabados

-1.08

ORDON

-1.08

GRAPHY

-1.07

 Gabriella

-1.06

貼り

-1.05

POSITIVE LOGITS

for

1.44

Then

1.41

 porque

1.16

 затем

1.13

пасибо

1.13

挹

1.13

 Hopefully

1.12

 länder

1.12

 потому

1.11

titudine

1.11

Activations Density 0.052%