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

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

Neuron 67 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.3900. Min Act: -0.1700

Data Index: 1168410 (C4 (Web Text))

Max Activating Token Index: 43

Text #1

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.3588. Min Act: -0.1700

Data Index: 531843 (C4 (Web Text))

Max Activating Token Index: 478

Text #2

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.3588. Min Act: -0.1700

Data Index: 599747 (C4 (Web Text))

Max Activating Token Index: 132

Text #3

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.2963. Min Act: -0.1700

Data Index: 238227 (C4 (Web Text))

Max Activating Token Index: 142

Text #4

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.2025. Min Act: -0.1700

Data Index: 945517 (C4 (Web Text))

Max Activating Token Index: 434

Text #5

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.2025. Min Act: -0.1700

Data Index: 926471 (C4 (Web Text))

Max Activating Token Index: 166

Text #6

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.1088. Min Act: -0.1700

Data Index: 617952 (C4 (Web Text))

Max Activating Token Index: 796

Text #7

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.0463. Min Act: -0.1700

Data Index: 1309861 (C4 (Web Text))

Max Activating Token Index: 879

Text #8

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.0775. Min Act: -0.1700

Data Index: 86884 (C4 (Web Text))

Max Activating Token Index: 615

Text #9

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.0463. Min Act: -0.1700

Data Index: 1281476 (C4 (Web Text))

Max Activating Token Index: 256

Text #10

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.0150. Min Act: -0.1700

Data Index: 636776 (C4 (Web Text))

Max Activating Token Index: 374

Text #11

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.0150. Min Act: -0.1700

Data Index: 1018028 (C4 (Web Text))

Max Activating Token Index: 781

Text #12

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.0150. Min Act: -0.1700

Data Index: 1342888 (C4 (Web Text))

Max Activating Token Index: 488

Text #13

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.0463. Min Act: -0.1700

Data Index: 25677 (C4 (Web Text))

Max Activating Token Index: 974

Text #14

Max Range: 5.3900. Min Range: -5.3900

Max Act: 4.9213. Min Act: -0.1700

Data Index: 1159743 (C4 (Web Text))

Max Activating Token Index: 633

Text #15

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.0150. Min Act: -0.1700

Data Index: 1283195 (C4 (Web Text))

Max Activating Token Index: 873

Text #16

Max Range: 5.3900. Min Range: -5.3900

Max Act: 5.0150. Min Act: -0.1700

Data Index: 504761 (C4 (Web Text))

Max Activating Token Index: 1011

Text #17

Max Range: 5.3900. Min Range: -5.3900

Max Act: 4.9838. Min Act: -0.1700

Data Index: 239545 (C4 (Web Text))

Max Activating Token Index: 947

Text #18

Max Range: 5.3900. Min Range: -5.3900

Max Act: 4.9525. Min Act: -0.1700

Data Index: 1002553 (C4 (Web Text))

Max Activating Token Index: 190

Text #19

Max Range: 5.3900. Min Range: -5.3900

Max Act: 4.9525. Min Act: -0.1700

Data Index: 883692 (C4 (Web Text))

Max Activating Token Index: 872