INDEX

Explanations

to identify, to put, to help

The neuron is highly responsive to the word “to” when it’s used to introduce an infinitive (i.e. purpose or action) in a sentence.

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

is

-1.70

for

-1.66

цене

-1.38

логу

-1.34

 để

-1.30

сей

-1.27

需要

-1.27

спор

-1.26

 gebre

-1.23

赑

-1.23

POSITIVE LOGITS

and

1.58

了

1.41

 aksesuar

1.39

 помогает

1.35

那时

1.34

もちろん

1.32

真實

1.30

 boast

1.28

也會

1.28

 akces

1.27

Activations Density 0.116%