INDEX

Explanations

The neuron consistently activates on verbs that denote movement or change of position—e.g. “stride,” “strides,” “shifts.”

stride, strides, shifts

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

évalu

-1.66

 them

-1.60

 Have

-1.55

酛

-1.52

If

-1.50

𖥧

-1.47

၃

-1.46

「

-1.41

 vultures

-1.40

浙江

-1.39

POSITIVE LOGITS

 frequentemente

2.03

 ambiti

1.84

 soooo

1.80

’

1.72

 intellect

1.72

,"

1.71

Ⴀ

1.64

<bos>

1.59

 vyrá

1.59

 sooo

1.58

Activations Density 0.002%