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

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

Neuron 2039 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: 3.6548. Min Range: -3.6548

Max Act: 3.6548. Min Act: -0.1700

Data Index: 495533 (C4 (Web Text))

Max Activating Token Index: 205

Text #1

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.4982. Min Act: -0.1700

Data Index: 207066 (C4 (Web Text))

Max Activating Token Index: 354

Text #2

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.4355. Min Act: -0.1700

Data Index: 920377 (C4 (Web Text))

Max Activating Token Index: 793

Text #3

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.4982. Min Act: -0.1700

Data Index: 304895 (C4 (Web Text))

Max Activating Token Index: 175

Text #4

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.4982. Min Act: -0.1700

Data Index: 179045 (C4 (Web Text))

Max Activating Token Index: 939

Text #5

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.4355. Min Act: -0.1700

Data Index: 339071 (C4 (Web Text))

Max Activating Token Index: 283

Text #6

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.4042. Min Act: -0.1700

Data Index: 936999 (C4 (Web Text))

Max Activating Token Index: 301

Text #7

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.4355. Min Act: -0.1700

Data Index: 307477 (C4 (Web Text))

Max Activating Token Index: 111

Text #8

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.3728. Min Act: -0.1700

Data Index: 560546 (C4 (Web Text))

Max Activating Token Index: 627

Text #9

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.4042. Min Act: -0.1700

Data Index: 102847 (C4 (Web Text))

Max Activating Token Index: 294

Text #10

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.3728. Min Act: -0.1700

Data Index: 1333885 (C4 (Web Text))

Max Activating Token Index: 649

Text #11

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.3571. Min Act: -0.1700

Data Index: 1250109 (C4 (Web Text))

Max Activating Token Index: 328

Text #12

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.3414. Min Act: -0.1700

Data Index: 973813 (C4 (Web Text))

Max Activating Token Index: 873

Text #13

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.2943. Min Act: -0.1700

Data Index: 766930 (C4 (Web Text))

Max Activating Token Index: 96

Text #14

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.2943. Min Act: -0.1700

Data Index: 332383 (C4 (Web Text))

Max Activating Token Index: 518

Text #15

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.3257. Min Act: -0.1700

Data Index: 1057753 (C4 (Web Text))

Max Activating Token Index: 42

Text #16

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.2943. Min Act: -0.1700

Data Index: 470569 (C4 (Web Text))

Max Activating Token Index: 632

Text #17

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.2943. Min Act: -0.1700

Data Index: 243012 (C4 (Web Text))

Max Activating Token Index: 229

Text #18

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.2943. Min Act: -0.1700

Data Index: 87889 (C4 (Web Text))

Max Activating Token Index: 404

Text #19

Max Range: 3.6548. Min Range: -3.6548

Max Act: 3.2629. Min Act: -0.1700

Data Index: 997993 (C4 (Web Text))

Max Activating Token Index: 195