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

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

Neuron 1893 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.4316. Min Act: -0.1700

Data Index: 1099500 (C4 (Web Text))

Max Activating Token Index: 298

Text #1

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.3691. Min Act: -0.1700

Data Index: 544934 (C4 (Web Text))

Max Activating Token Index: 953

Text #2

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.2441. Min Act: -0.1700

Data Index: 1081194 (C4 (Web Text))

Max Activating Token Index: 925

Text #3

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.2128. Min Act: -0.1700

Data Index: 573 (C4 (Web Text))

Max Activating Token Index: 510

Text #4

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.2441. Min Act: -0.1700

Data Index: 605304 (C4 (Web Text))

Max Activating Token Index: 568

Text #5

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.2441. Min Act: -0.1700

Data Index: 463529 (C4 (Web Text))

Max Activating Token Index: 465

Text #6

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.1816. Min Act: -0.1700

Data Index: 412864 (C4 (Web Text))

Max Activating Token Index: 136

Text #7

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.1503. Min Act: -0.1700

Data Index: 686971 (C4 (Web Text))

Max Activating Token Index: 1014

Text #8

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.1816. Min Act: -0.1700

Data Index: 1189170 (C4 (Web Text))

Max Activating Token Index: 328

Text #9

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.1503. Min Act: -0.1700

Data Index: 808347 (C4 (Web Text))

Max Activating Token Index: 776

Text #10

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.1191. Min Act: -0.1700

Data Index: 635820 (C4 (Web Text))

Max Activating Token Index: 718

Text #11

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.1503. Min Act: -0.1700

Data Index: 349394 (C4 (Web Text))

Max Activating Token Index: 167

Text #12

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.0878. Min Act: -0.1700

Data Index: 188682 (C4 (Web Text))

Max Activating Token Index: 955

Text #13

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.0878. Min Act: -0.1700

Data Index: 840344 (C4 (Web Text))

Max Activating Token Index: 366

Text #14

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.0878. Min Act: -0.1700

Data Index: 1170483 (C4 (Web Text))

Max Activating Token Index: 634

Text #15

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.0566. Min Act: -0.1700

Data Index: 665974 (C4 (Web Text))

Max Activating Token Index: 786

Text #16

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.0253. Min Act: -0.1700

Data Index: 656776 (C4 (Web Text))

Max Activating Token Index: 705

Text #17

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.0566. Min Act: -0.1700

Data Index: 380256 (C4 (Web Text))

Max Activating Token Index: 518

Text #18

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.0253. Min Act: -0.1700

Data Index: 1084851 (C4 (Web Text))

Max Activating Token Index: 914

Text #19

Max Range: 5.4316. Min Range: -5.4316

Max Act: 5.0253. Min Act: -0.1700

Data Index: 467128 (C4 (Web Text))

Max Activating Token Index: 458