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

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

Neuron 1467 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.5981. Min Act: -0.1700

Data Index: 1319235 (C4 (Web Text))

Max Activating Token Index: 682

Text #1

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4728. Min Act: -0.1700

Data Index: 819551 (C4 (Web Text))

Max Activating Token Index: 818

Text #2

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.5355. Min Act: -0.1700

Data Index: 1338206 (C4 (Web Text))

Max Activating Token Index: 870

Text #3

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.5041. Min Act: -0.1700

Data Index: 39999 (C4 (Web Text))

Max Activating Token Index: 776

Text #4

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4571. Min Act: -0.1700

Data Index: 445992 (C4 (Web Text))

Max Activating Token Index: 823

Text #5

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.5041. Min Act: -0.1700

Data Index: 1235186 (C4 (Web Text))

Max Activating Token Index: 678

Text #6

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.5041. Min Act: -0.1700

Data Index: 884974 (C4 (Web Text))

Max Activating Token Index: 717

Text #7

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4728. Min Act: -0.1700

Data Index: 1595419 (Python Code)

Max Activating Token Index: 847

Text #8

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4415. Min Act: -0.1700

Data Index: 280029 (C4 (Web Text))

Max Activating Token Index: 859

Text #9

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4101. Min Act: -0.1700

Data Index: 39731 (C4 (Web Text))

Max Activating Token Index: 274

Text #10

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4728. Min Act: -0.1700

Data Index: 1081951 (C4 (Web Text))

Max Activating Token Index: 308

Text #11

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4101. Min Act: -0.1700

Data Index: 67799 (C4 (Web Text))

Max Activating Token Index: 870

Text #12

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4258. Min Act: -0.1700

Data Index: 1317748 (C4 (Web Text))

Max Activating Token Index: 167

Text #13

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4415. Min Act: -0.1700

Data Index: 1043897 (C4 (Web Text))

Max Activating Token Index: 469

Text #14

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4571. Min Act: -0.1700

Data Index: 1055633 (C4 (Web Text))

Max Activating Token Index: 870

Text #15

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4258. Min Act: -0.1700

Data Index: 1557909 (Python Code)

Max Activating Token Index: 885

Text #16

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4101. Min Act: -0.1700

Data Index: 919025 (C4 (Web Text))

Max Activating Token Index: 887

Text #17

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.4571. Min Act: -0.1700

Data Index: 868663 (C4 (Web Text))

Max Activating Token Index: 445

Text #18

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.3787. Min Act: -0.1700

Data Index: 1002915 (C4 (Web Text))

Max Activating Token Index: 432

Text #19

Max Range: 3.5981. Min Range: -3.5981

Max Act: 3.3473. Min Act: -0.1700

Data Index: 1097936 (C4 (Web Text))

Max Activating Token Index: 388