INDEX

Explanations

actions followed by 'to' or objects

The neuron specifically detects occurrences of the word “playing.”

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

){$

-1.35

archical

-1.32

 část

-1.30

aurait

-1.30

 男性

-1.24

包括

-1.24

 soldadura

-1.23

tolua

-1.22

 brauch

-1.21

によって

-1.20

POSITIVE LOGITS

๘

1.27

Now

1.24

 this

1.23

with

1.21

 these

1.20

To

1.18

 From

1.16

Alongside

1.13

you

1.12

 actualmente

1.10

Activations Density 0.073%