INDEX

Explanations

say something that

The neuron activates on occurrences of reporting verbs like “say,” “says,” or “said.”

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

tiéndose

-1.29

ratory

-1.28

oczes

-1.26

///</

-1.24

ਗ਼

-1.21

/>

-1.19

preneurship

-1.18

)=>{

-1.18

'])

-1.16

</h4>

-1.16

POSITIVE LOGITS

 that

3.22

it

2.33

 they

2.25

 there

2.19

you

1.47

we

1.45

to

1.44

：……

1.41

 things

1.31

 this

1.30

Activations Density 0.032%