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

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

Neuron 679 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.9571. Min Act: -0.1700

Data Index: 72483 (C4 (Web Text))

Max Activating Token Index: 74

Text #1

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.8164. Min Act: -0.1700

Data Index: 1064914 (C4 (Web Text))

Max Activating Token Index: 160

Text #2

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.7538. Min Act: -0.1700

Data Index: 1012609 (C4 (Web Text))

Max Activating Token Index: 10

Text #3

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.5817. Min Act: -0.1700

Data Index: 1378143 (Python Code)

Max Activating Token Index: 9

Text #4

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.6130. Min Act: -0.1700

Data Index: 665089 (C4 (Web Text))

Max Activating Token Index: 525

Text #5

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.5660. Min Act: -0.1700

Data Index: 1059111 (C4 (Web Text))

Max Activating Token Index: 644

Text #6

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.5817. Min Act: -0.1700

Data Index: 517956 (C4 (Web Text))

Max Activating Token Index: 109

Text #7

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.5660. Min Act: -0.1700

Data Index: 1195654 (C4 (Web Text))

Max Activating Token Index: 541

Text #8

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.5034. Min Act: -0.1700

Data Index: 891331 (C4 (Web Text))

Max Activating Token Index: 188

Text #9

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4877. Min Act: -0.1700

Data Index: 741998 (C4 (Web Text))

Max Activating Token Index: 965

Text #10

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4250. Min Act: -0.1700

Data Index: 434677 (C4 (Web Text))

Max Activating Token Index: 914

Text #11

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4563. Min Act: -0.1700

Data Index: 856240 (C4 (Web Text))

Max Activating Token Index: 809

Text #12

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4563. Min Act: -0.1700

Data Index: 985277 (C4 (Web Text))

Max Activating Token Index: 82

Text #13

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4407. Min Act: -0.1700

Data Index: 450921 (C4 (Web Text))

Max Activating Token Index: 171

Text #14

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4093. Min Act: -0.1700

Data Index: 789131 (C4 (Web Text))

Max Activating Token Index: 74

Text #15

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4407. Min Act: -0.1700

Data Index: 1295550 (C4 (Web Text))

Max Activating Token Index: 728

Text #16

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4093. Min Act: -0.1700

Data Index: 880258 (C4 (Web Text))

Max Activating Token Index: 57

Text #17

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4720. Min Act: -0.1700

Data Index: 137666 (C4 (Web Text))

Max Activating Token Index: 431

Text #18

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4407. Min Act: -0.1700

Data Index: 1175580 (C4 (Web Text))

Max Activating Token Index: 50

Text #19

Max Range: 3.9571. Min Range: -3.9571

Max Act: 3.4250. Min Act: -0.1700

Data Index: 1399671 (Python Code)

Max Activating Token Index: 5