Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1775 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.5297. Min Act: -0.1700

Data Index: 505651 (C4 (Web Text))

Max Activating Token Index: 387

Text #1

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.5609. Min Act: -0.1700

Data Index: 1300614 (C4 (Web Text))

Max Activating Token Index: 751

Text #2

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.4359. Min Act: -0.1700

Data Index: 201650 (C4 (Web Text))

Max Activating Token Index: 999

Text #3

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.5297. Min Act: -0.1700

Data Index: 477202 (C4 (Web Text))

Max Activating Token Index: 901

Text #4

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.3422. Min Act: -0.1700

Data Index: 198455 (C4 (Web Text))

Max Activating Token Index: 352

Text #5

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.3109. Min Act: -0.1700

Data Index: 1297111 (C4 (Web Text))

Max Activating Token Index: 437

Text #6

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.3109. Min Act: -0.1700

Data Index: 968859 (C4 (Web Text))

Max Activating Token Index: 351

Text #7

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.3109. Min Act: -0.1700

Data Index: 235061 (C4 (Web Text))

Max Activating Token Index: 780

Text #8

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.3109. Min Act: -0.1700

Data Index: 884088 (C4 (Web Text))

Max Activating Token Index: 21

Text #9

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.3109. Min Act: -0.1700

Data Index: 1033446 (C4 (Web Text))

Max Activating Token Index: 551

Text #10

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.2797. Min Act: -0.1699

Data Index: 804304 (C4 (Web Text))

Max Activating Token Index: 442

Text #11

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.2172. Min Act: -0.1700

Data Index: 105232 (C4 (Web Text))

Max Activating Token Index: 410

Text #12

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.2484. Min Act: -0.1700

Data Index: 170921 (C4 (Web Text))

Max Activating Token Index: 960

Text #13

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.1859. Min Act: -0.1699

Data Index: 257683 (C4 (Web Text))

Max Activating Token Index: 887

Text #14

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.2172. Min Act: -0.1700

Data Index: 750844 (C4 (Web Text))

Max Activating Token Index: 686

Text #15

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.1547. Min Act: -0.1700

Data Index: 578146 (C4 (Web Text))

Max Activating Token Index: 178

Text #16

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.1859. Min Act: -0.1700

Data Index: 1305878 (C4 (Web Text))

Max Activating Token Index: 69

Text #17

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.0922. Min Act: -0.1700

Data Index: 837550 (C4 (Web Text))

Max Activating Token Index: 840

Text #18

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.0922. Min Act: -0.1700

Data Index: 19850 (C4 (Web Text))

Max Activating Token Index: 547

Text #19

Max Range: 5.5609. Min Range: -5.5609

Max Act: 5.1234. Min Act: -0.1700

Data Index: 1322148 (C4 (Web Text))

Max Activating Token Index: 857