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

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

Neuron 350 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.5707. Min Act: -0.1700

Data Index: 27994 (C4 (Web Text))

Max Activating Token Index: 355

Text #1

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.5707. Min Act: -0.1700

Data Index: 1202917 (C4 (Web Text))

Max Activating Token Index: 693

Text #2

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.5550. Min Act: -0.1700

Data Index: 1035612 (C4 (Web Text))

Max Activating Token Index: 525

Text #3

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4924. Min Act: -0.1700

Data Index: 1206593 (C4 (Web Text))

Max Activating Token Index: 305

Text #4

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4767. Min Act: -0.1700

Data Index: 394891 (C4 (Web Text))

Max Activating Token Index: 791

Text #5

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4924. Min Act: -0.1700

Data Index: 1356086 (C4 (Web Text))

Max Activating Token Index: 154

Text #6

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.5237. Min Act: -0.1700

Data Index: 350399 (C4 (Web Text))

Max Activating Token Index: 202

Text #7

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4297. Min Act: -0.1700

Data Index: 603174 (C4 (Web Text))

Max Activating Token Index: 953

Text #8

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4453. Min Act: -0.1700

Data Index: 527277 (C4 (Web Text))

Max Activating Token Index: 316

Text #9

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.3826. Min Act: -0.1700

Data Index: 284769 (C4 (Web Text))

Max Activating Token Index: 205

Text #10

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4297. Min Act: -0.1700

Data Index: 1018523 (C4 (Web Text))

Max Activating Token Index: 261

Text #11

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4140. Min Act: -0.1700

Data Index: 326564 (C4 (Web Text))

Max Activating Token Index: 351

Text #12

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4140. Min Act: -0.1700

Data Index: 1335019 (C4 (Web Text))

Max Activating Token Index: 771

Text #13

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4140. Min Act: -0.1700

Data Index: 1069154 (C4 (Web Text))

Max Activating Token Index: 958

Text #14

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4297. Min Act: -0.1700

Data Index: 1333377 (C4 (Web Text))

Max Activating Token Index: 653

Text #15

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.4140. Min Act: -0.1700

Data Index: 1127885 (C4 (Web Text))

Max Activating Token Index: 448

Text #16

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.3983. Min Act: -0.1700

Data Index: 28494 (C4 (Web Text))

Max Activating Token Index: 358

Text #17

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.3826. Min Act: -0.1700

Data Index: 1231200 (C4 (Web Text))

Max Activating Token Index: 149

Text #18

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.3826. Min Act: -0.1700

Data Index: 1000142 (C4 (Web Text))

Max Activating Token Index: 919

Text #19

Max Range: 3.5707. Min Range: -3.5707

Max Act: 3.3512. Min Act: -0.1700

Data Index: 259873 (C4 (Web Text))

Max Activating Token Index: 98