Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1039 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.0088. Min Range: -3.0088

Max Act: 3.0088. Min Act: -0.1700

Data Index: 1222447 (C4 (Web Text))

Max Activating Token Index: 745

Text #1

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.9614. Min Act: -0.1700

Data Index: 636794 (C4 (Web Text))

Max Activating Token Index: 685

Text #2

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.9614. Min Act: -0.1700

Data Index: 956334 (C4 (Web Text))

Max Activating Token Index: 690

Text #3

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.9297. Min Act: -0.1700

Data Index: 256886 (C4 (Web Text))

Max Activating Token Index: 351

Text #4

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.9138. Min Act: -0.1700

Data Index: 1210724 (C4 (Web Text))

Max Activating Token Index: 219

Text #5

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8980. Min Act: -0.1700

Data Index: 810467 (C4 (Web Text))

Max Activating Token Index: 76

Text #6

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8980. Min Act: -0.1700

Data Index: 1302388 (C4 (Web Text))

Max Activating Token Index: 372

Text #7

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8980. Min Act: -0.1700

Data Index: 879870 (C4 (Web Text))

Max Activating Token Index: 730

Text #8

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8980. Min Act: -0.1700

Data Index: 1358269 (C4 (Web Text))

Max Activating Token Index: 380

Text #9

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.9138. Min Act: -0.1700

Data Index: 471419 (C4 (Web Text))

Max Activating Token Index: 574

Text #10

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.9138. Min Act: -0.1700

Data Index: 951778 (C4 (Web Text))

Max Activating Token Index: 874

Text #11

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8821. Min Act: -0.1700

Data Index: 1445819 (Python Code)

Max Activating Token Index: 372

Text #12

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8980. Min Act: -0.1700

Data Index: 1000343 (C4 (Web Text))

Max Activating Token Index: 362

Text #13

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8821. Min Act: -0.1700

Data Index: 750679 (C4 (Web Text))

Max Activating Token Index: 899

Text #14

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8821. Min Act: -0.1700

Data Index: 1198195 (C4 (Web Text))

Max Activating Token Index: 765

Text #15

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8821. Min Act: -0.1700

Data Index: 94957 (C4 (Web Text))

Max Activating Token Index: 492

Text #16

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8821. Min Act: -0.1700

Data Index: 911067 (C4 (Web Text))

Max Activating Token Index: 416

Text #17

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8503. Min Act: -0.1700

Data Index: 663394 (C4 (Web Text))

Max Activating Token Index: 374

Text #18

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8503. Min Act: -0.1700

Data Index: 1113984 (C4 (Web Text))

Max Activating Token Index: 398

Text #19

Max Range: 3.0088. Min Range: -3.0088

Max Act: 2.8503. Min Act: -0.1700

Data Index: 538785 (C4 (Web Text))

Max Activating Token Index: 366