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

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

Neuron 578 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: 3.4550. Min Range: -3.4550

Max Act: 3.4550. Min Act: -0.1700

Data Index: 80452 (C4 (Web Text))

Max Activating Token Index: 726

Text #1

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.4550. Min Act: -0.1700

Data Index: 1129120 (C4 (Web Text))

Max Activating Token Index: 157

Text #2

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3923. Min Act: -0.1700

Data Index: 2719 (C4 (Web Text))

Max Activating Token Index: 109

Text #3

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.4080. Min Act: -0.1700

Data Index: 603905 (C4 (Web Text))

Max Activating Token Index: 153

Text #4

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.4080. Min Act: -0.1700

Data Index: 110497 (C4 (Web Text))

Max Activating Token Index: 859

Text #5

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.4236. Min Act: -0.1700

Data Index: 685212 (C4 (Web Text))

Max Activating Token Index: 298

Text #6

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3766. Min Act: -0.1700

Data Index: 188355 (C4 (Web Text))

Max Activating Token Index: 11

Text #7

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3766. Min Act: -0.1700

Data Index: 525393 (C4 (Web Text))

Max Activating Token Index: 940

Text #8

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3766. Min Act: -0.1700

Data Index: 1318296 (C4 (Web Text))

Max Activating Token Index: 470

Text #9

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3766. Min Act: -0.1700

Data Index: 990692 (C4 (Web Text))

Max Activating Token Index: 259

Text #10

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3452. Min Act: -0.1700

Data Index: 640988 (C4 (Web Text))

Max Activating Token Index: 578

Text #11

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3452. Min Act: -0.1700

Data Index: 315125 (C4 (Web Text))

Max Activating Token Index: 133

Text #12

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3452. Min Act: -0.1700

Data Index: 781721 (C4 (Web Text))

Max Activating Token Index: 1020

Text #13

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3295. Min Act: -0.1700

Data Index: 282016 (C4 (Web Text))

Max Activating Token Index: 547

Text #14

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3452. Min Act: -0.1700

Data Index: 982845 (C4 (Web Text))

Max Activating Token Index: 625

Text #15

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3609. Min Act: -0.1700

Data Index: 633521 (C4 (Web Text))

Max Activating Token Index: 891

Text #16

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3295. Min Act: -0.1700

Data Index: 574860 (C4 (Web Text))

Max Activating Token Index: 569

Text #17

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3138. Min Act: -0.1700

Data Index: 625534 (C4 (Web Text))

Max Activating Token Index: 214

Text #18

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3295. Min Act: -0.1700

Data Index: 537277 (C4 (Web Text))

Max Activating Token Index: 234

Text #19

Max Range: 3.4550. Min Range: -3.4550

Max Act: 3.3138. Min Act: -0.1700

Data Index: 67156 (C4 (Web Text))

Max Activating Token Index: 566