Model: GELU Model: 4 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 464 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.8973. Min Act: -0.1700

Data Index: 125367 (C4 (Web Text))

Max Activating Token Index: 552

Text #1

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.8660. Min Act: -0.1700

Data Index: 741541 (C4 (Web Text))

Max Activating Token Index: 816

Text #2

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.8348. Min Act: -0.1700

Data Index: 174628 (C4 (Web Text))

Max Activating Token Index: 799

Text #3

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7722. Min Act: -0.1700

Data Index: 1179283 (C4 (Web Text))

Max Activating Token Index: 517

Text #4

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7722. Min Act: -0.1700

Data Index: 1072991 (C4 (Web Text))

Max Activating Token Index: 445

Text #5

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7409. Min Act: -0.1700

Data Index: 1009987 (C4 (Web Text))

Max Activating Token Index: 754

Text #6

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7096. Min Act: -0.1700

Data Index: 1318872 (C4 (Web Text))

Max Activating Token Index: 1017

Text #7

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7409. Min Act: -0.1700

Data Index: 595109 (C4 (Web Text))

Max Activating Token Index: 567

Text #8

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7722. Min Act: -0.1700

Data Index: 97957 (C4 (Web Text))

Max Activating Token Index: 715

Text #9

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7096. Min Act: -0.1700

Data Index: 1133527 (C4 (Web Text))

Max Activating Token Index: 803

Text #10

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7096. Min Act: -0.1700

Data Index: 507029 (C4 (Web Text))

Max Activating Token Index: 952

Text #11

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7096. Min Act: -0.1700

Data Index: 482688 (C4 (Web Text))

Max Activating Token Index: 37

Text #12

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.6783. Min Act: -0.1700

Data Index: 1208473 (C4 (Web Text))

Max Activating Token Index: 768

Text #13

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.6627. Min Act: -0.1700

Data Index: 716044 (C4 (Web Text))

Max Activating Token Index: 852

Text #14

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.7096. Min Act: -0.1700

Data Index: 35644 (C4 (Web Text))

Max Activating Token Index: 263

Text #15

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.6783. Min Act: -0.1700

Data Index: 185741 (C4 (Web Text))

Max Activating Token Index: 381

Text #16

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.6783. Min Act: -0.1700

Data Index: 449966 (C4 (Web Text))

Max Activating Token Index: 989

Text #17

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.6783. Min Act: -0.1700

Data Index: 1103304 (C4 (Web Text))

Max Activating Token Index: 27

Text #18

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.6627. Min Act: -0.1700

Data Index: 640062 (C4 (Web Text))

Max Activating Token Index: 900

Text #19

Max Range: 3.8973. Min Range: -3.8973

Max Act: 3.6783. Min Act: -0.1700

Data Index: 653388 (C4 (Web Text))

Max Activating Token Index: 371