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

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

Neuron 1460 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2154. Min Range: -3.2154

Max Act: 3.2154. Min Act: -0.1700

Data Index: 881649 (C4 (Web Text))

Max Activating Token Index: 5

Text #1

Max Range: 3.2154. Min Range: -3.2154

Max Act: 3.0262. Min Act: -0.1700

Data Index: 653442 (C4 (Web Text))

Max Activating Token Index: 969

Text #2

Max Range: 3.2154. Min Range: -3.2154

Max Act: 3.0262. Min Act: -0.1700

Data Index: 979461 (C4 (Web Text))

Max Activating Token Index: 9

Text #3

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9946. Min Act: -0.1700

Data Index: 549112 (C4 (Web Text))

Max Activating Token Index: 376

Text #4

Max Range: 3.2154. Min Range: -3.2154

Max Act: 3.0104. Min Act: -0.1700

Data Index: 1342195 (C4 (Web Text))

Max Activating Token Index: 119

Text #5

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9313. Min Act: -0.1700

Data Index: 295163 (C4 (Web Text))

Max Activating Token Index: 1020

Text #6

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9788. Min Act: -0.1700

Data Index: 837071 (C4 (Web Text))

Max Activating Token Index: 668

Text #7

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9471. Min Act: -0.1700

Data Index: 1381098 (Python Code)

Max Activating Token Index: 5

Text #8

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9471. Min Act: -0.1700

Data Index: 238243 (C4 (Web Text))

Max Activating Token Index: 356

Text #9

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9471. Min Act: -0.1700

Data Index: 515806 (C4 (Web Text))

Max Activating Token Index: 555

Text #10

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9630. Min Act: -0.1700

Data Index: 831721 (C4 (Web Text))

Max Activating Token Index: 271

Text #11

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9155. Min Act: -0.1700

Data Index: 376443 (C4 (Web Text))

Max Activating Token Index: 77

Text #12

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9155. Min Act: -0.1700

Data Index: 152641 (C4 (Web Text))

Max Activating Token Index: 998

Text #13

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.8996. Min Act: -0.1700

Data Index: 199859 (C4 (Web Text))

Max Activating Token Index: 501

Text #14

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.9155. Min Act: -0.1700

Data Index: 1006551 (C4 (Web Text))

Max Activating Token Index: 667

Text #15

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.8679. Min Act: -0.1700

Data Index: 1400384 (Python Code)

Max Activating Token Index: 710

Text #16

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.8996. Min Act: -0.1700

Data Index: 1597006 (Python Code)

Max Activating Token Index: 819

Text #17

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.8837. Min Act: -0.1700

Data Index: 847390 (C4 (Web Text))

Max Activating Token Index: 889

Text #18

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.8996. Min Act: -0.1700

Data Index: 1077387 (C4 (Web Text))

Max Activating Token Index: 17

Text #19

Max Range: 3.2154. Min Range: -3.2154

Max Act: 2.8996. Min Act: -0.1700

Data Index: 599368 (C4 (Web Text))

Max Activating Token Index: 55