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

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

Neuron 462 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: 3.6939. Min Range: -3.6939

Max Act: 3.6939. Min Act: -0.1700

Data Index: 176330 (C4 (Web Text))

Max Activating Token Index: 427

Text #1

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.6000. Min Act: -0.1700

Data Index: 998010 (C4 (Web Text))

Max Activating Token Index: 891

Text #2

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.5687. Min Act: -0.1700

Data Index: 869352 (C4 (Web Text))

Max Activating Token Index: 699

Text #3

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.5843. Min Act: -0.1700

Data Index: 319109 (C4 (Web Text))

Max Activating Token Index: 310

Text #4

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4747. Min Act: -0.1700

Data Index: 31882 (C4 (Web Text))

Max Activating Token Index: 8

Text #5

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.5060. Min Act: -0.1700

Data Index: 831272 (C4 (Web Text))

Max Activating Token Index: 419

Text #6

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4747. Min Act: -0.1700

Data Index: 585102 (C4 (Web Text))

Max Activating Token Index: 937

Text #7

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4747. Min Act: -0.1700

Data Index: 864406 (C4 (Web Text))

Max Activating Token Index: 563

Text #8

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4903. Min Act: -0.1700

Data Index: 474295 (C4 (Web Text))

Max Activating Token Index: 908

Text #9

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4747. Min Act: -0.1700

Data Index: 863482 (C4 (Web Text))

Max Activating Token Index: 624

Text #10

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4747. Min Act: -0.1700

Data Index: 252430 (C4 (Web Text))

Max Activating Token Index: 733

Text #11

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.5060. Min Act: -0.1700

Data Index: 1258 (C4 (Web Text))

Max Activating Token Index: 382

Text #12

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4747. Min Act: -0.1700

Data Index: 495152 (C4 (Web Text))

Max Activating Token Index: 789

Text #13

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4747. Min Act: -0.1700

Data Index: 854505 (C4 (Web Text))

Max Activating Token Index: 144

Text #14

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4590. Min Act: -0.1700

Data Index: 623169 (C4 (Web Text))

Max Activating Token Index: 471

Text #15

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4590. Min Act: -0.1700

Data Index: 1133098 (C4 (Web Text))

Max Activating Token Index: 793

Text #16

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4590. Min Act: -0.1700

Data Index: 594613 (C4 (Web Text))

Max Activating Token Index: 668

Text #17

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4120. Min Act: -0.1700

Data Index: 73883 (C4 (Web Text))

Max Activating Token Index: 746

Text #18

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4433. Min Act: -0.1700

Data Index: 709131 (C4 (Web Text))

Max Activating Token Index: 542

Text #19

Max Range: 3.6939. Min Range: -3.6939

Max Act: 3.4276. Min Act: -0.1700

Data Index: 296002 (C4 (Web Text))

Max Activating Token Index: 642