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

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

Neuron 999 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.4595. Min Act: -0.1700

Data Index: 758260 (C4 (Web Text))

Max Activating Token Index: 1008

Text #1

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3968. Min Act: -0.1700

Data Index: 359283 (C4 (Web Text))

Max Activating Token Index: 909

Text #2

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.4281. Min Act: -0.1700

Data Index: 1409390 (Python Code)

Max Activating Token Index: 909

Text #3

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.4595. Min Act: -0.1700

Data Index: 803378 (C4 (Web Text))

Max Activating Token Index: 716

Text #4

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3968. Min Act: -0.1700

Data Index: 1341496 (C4 (Web Text))

Max Activating Token Index: 927

Text #5

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3968. Min Act: -0.1700

Data Index: 1582345 (Python Code)

Max Activating Token Index: 355

Text #6

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3968. Min Act: -0.1700

Data Index: 1277772 (C4 (Web Text))

Max Activating Token Index: 289

Text #7

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3654. Min Act: -0.1700

Data Index: 370700 (C4 (Web Text))

Max Activating Token Index: 498

Text #8

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3968. Min Act: -0.1700

Data Index: 185525 (C4 (Web Text))

Max Activating Token Index: 543

Text #9

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3968. Min Act: -0.1700

Data Index: 1177261 (C4 (Web Text))

Max Activating Token Index: 112

Text #10

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3654. Min Act: -0.1700

Data Index: 635267 (C4 (Web Text))

Max Activating Token Index: 942

Text #11

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3968. Min Act: -0.1700

Data Index: 277915 (C4 (Web Text))

Max Activating Token Index: 428

Text #12

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3340. Min Act: -0.1700

Data Index: 267706 (C4 (Web Text))

Max Activating Token Index: 992

Text #13

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.2869. Min Act: -0.1700

Data Index: 180860 (C4 (Web Text))

Max Activating Token Index: 568

Text #14

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3340. Min Act: -0.1700

Data Index: 741806 (C4 (Web Text))

Max Activating Token Index: 492

Text #15

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3340. Min Act: -0.1700

Data Index: 247470 (C4 (Web Text))

Max Activating Token Index: 50

Text #16

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3183. Min Act: -0.1700

Data Index: 482246 (C4 (Web Text))

Max Activating Token Index: 166

Text #17

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3340. Min Act: -0.1700

Data Index: 599717 (C4 (Web Text))

Max Activating Token Index: 432

Text #18

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3183. Min Act: -0.1700

Data Index: 1002905 (C4 (Web Text))

Max Activating Token Index: 1004

Text #19

Max Range: 3.4595. Min Range: -3.4595

Max Act: 3.3340. Min Act: -0.1700

Data Index: 1264284 (C4 (Web Text))

Max Activating Token Index: 293