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

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

Neuron 320 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: 4.1009. Min Range: -4.1009

Max Act: 4.1009. Min Act: -0.1700

Data Index: 496907 (C4 (Web Text))

Max Activating Token Index: 33

Text #1

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9759. Min Act: -0.1700

Data Index: 463722 (C4 (Web Text))

Max Activating Token Index: 121

Text #2

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9759. Min Act: -0.1700

Data Index: 604412 (C4 (Web Text))

Max Activating Token Index: 94

Text #3

Max Range: 4.1009. Min Range: -4.1009

Max Act: 4.0071. Min Act: -0.1700

Data Index: 988500 (C4 (Web Text))

Max Activating Token Index: 995

Text #4

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9759. Min Act: -0.1700

Data Index: 1335933 (C4 (Web Text))

Max Activating Token Index: 15

Text #5

Max Range: 4.1009. Min Range: -4.1009

Max Act: 4.1009. Min Act: -0.1700

Data Index: 961140 (C4 (Web Text))

Max Activating Token Index: 422

Text #6

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.8195. Min Act: -0.1700

Data Index: 559377 (C4 (Web Text))

Max Activating Token Index: 223

Text #7

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9446. Min Act: -0.1700

Data Index: 307955 (C4 (Web Text))

Max Activating Token Index: 448

Text #8

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9133. Min Act: -0.1700

Data Index: 516151 (C4 (Web Text))

Max Activating Token Index: 101

Text #9

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9133. Min Act: -0.1700

Data Index: 402105 (C4 (Web Text))

Max Activating Token Index: 278

Text #10

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9759. Min Act: -0.1700

Data Index: 529696 (C4 (Web Text))

Max Activating Token Index: 670

Text #11

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9446. Min Act: -0.1700

Data Index: 622928 (C4 (Web Text))

Max Activating Token Index: 52

Text #12

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9759. Min Act: -0.1700

Data Index: 37205 (C4 (Web Text))

Max Activating Token Index: 149

Text #13

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9133. Min Act: -0.1700

Data Index: 566621 (C4 (Web Text))

Max Activating Token Index: 489

Text #14

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.8821. Min Act: -0.1700

Data Index: 440096 (C4 (Web Text))

Max Activating Token Index: 80

Text #15

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9133. Min Act: -0.1700

Data Index: 1026937 (C4 (Web Text))

Max Activating Token Index: 1022

Text #16

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9446. Min Act: -0.1700

Data Index: 1042035 (C4 (Web Text))

Max Activating Token Index: 128

Text #17

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.9133. Min Act: -0.1700

Data Index: 507097 (C4 (Web Text))

Max Activating Token Index: 42

Text #18

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.8821. Min Act: -0.1700

Data Index: 28411 (C4 (Web Text))

Max Activating Token Index: 55

Text #19

Max Range: 4.1009. Min Range: -4.1009

Max Act: 3.8821. Min Act: -0.1700

Data Index: 676056 (C4 (Web Text))

Max Activating Token Index: 552