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

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

Neuron 1801 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.5873. Min Act: -0.1700

Data Index: 1084278 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.5560. Min Act: -0.1700

Data Index: 986813 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.4933. Min Act: -0.1700

Data Index: 347328 (C4 (Web Text))

Max Activating Token Index: 539

Text #3

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.4776. Min Act: -0.1700

Data Index: 947278 (C4 (Web Text))

Max Activating Token Index: 158

Text #4

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.4776. Min Act: -0.1700

Data Index: 909966 (C4 (Web Text))

Max Activating Token Index: 189

Text #5

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.4463. Min Act: -0.1700

Data Index: 413314 (C4 (Web Text))

Max Activating Token Index: 468

Text #6

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.4149. Min Act: -0.1700

Data Index: 1020911 (C4 (Web Text))

Max Activating Token Index: 972

Text #7

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.4149. Min Act: -0.1700

Data Index: 1217699 (C4 (Web Text))

Max Activating Token Index: 635

Text #8

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3992. Min Act: -0.1700

Data Index: 65151 (C4 (Web Text))

Max Activating Token Index: 335

Text #9

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.4149. Min Act: -0.1700

Data Index: 28671 (C4 (Web Text))

Max Activating Token Index: 570

Text #10

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3835. Min Act: -0.1700

Data Index: 46004 (C4 (Web Text))

Max Activating Token Index: 232

Text #11

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3835. Min Act: -0.1700

Data Index: 258858 (C4 (Web Text))

Max Activating Token Index: 946

Text #12

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3835. Min Act: -0.1700

Data Index: 1231217 (C4 (Web Text))

Max Activating Token Index: 873

Text #13

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3992. Min Act: -0.1700

Data Index: 476244 (C4 (Web Text))

Max Activating Token Index: 427

Text #14

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3992. Min Act: -0.1700

Data Index: 1450373 (Python Code)

Max Activating Token Index: 1023

Text #15

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3679. Min Act: -0.1700

Data Index: 633424 (C4 (Web Text))

Max Activating Token Index: 472

Text #16

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3679. Min Act: -0.1700

Data Index: 1297676 (C4 (Web Text))

Max Activating Token Index: 360

Text #17

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3679. Min Act: -0.1700

Data Index: 794752 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3522. Min Act: -0.1700

Data Index: 1061972 (C4 (Web Text))

Max Activating Token Index: 363

Text #19

Max Range: 3.5873. Min Range: -3.5873

Max Act: 3.3835. Min Act: -0.1700

Data Index: 826794 (C4 (Web Text))

Max Activating Token Index: 926