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

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

Neuron 78 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.3953. Min Act: -0.1700

Data Index: 711243 (C4 (Web Text))

Max Activating Token Index: 901

Text #1

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.3016. Min Act: -0.1700

Data Index: 510326 (C4 (Web Text))

Max Activating Token Index: 849

Text #2

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.3016. Min Act: -0.1700

Data Index: 39331 (C4 (Web Text))

Max Activating Token Index: 918

Text #3

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.2703. Min Act: -0.1700

Data Index: 561571 (C4 (Web Text))

Max Activating Token Index: 909

Text #4

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.2391. Min Act: -0.1700

Data Index: 346950 (C4 (Web Text))

Max Activating Token Index: 490

Text #5

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.2391. Min Act: -0.1700

Data Index: 1084927 (C4 (Web Text))

Max Activating Token Index: 279

Text #6

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.1141. Min Act: -0.1700

Data Index: 568665 (C4 (Web Text))

Max Activating Token Index: 597

Text #7

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.0828. Min Act: -0.1700

Data Index: 1211265 (C4 (Web Text))

Max Activating Token Index: 698

Text #8

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.0828. Min Act: -0.1700

Data Index: 780732 (C4 (Web Text))

Max Activating Token Index: 260

Text #9

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.0516. Min Act: -0.1700

Data Index: 524550 (C4 (Web Text))

Max Activating Token Index: 932

Text #10

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.0203. Min Act: -0.1700

Data Index: 630408 (C4 (Web Text))

Max Activating Token Index: 131

Text #11

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.0203. Min Act: -0.1700

Data Index: 1149169 (C4 (Web Text))

Max Activating Token Index: 295

Text #12

Max Range: 5.3953. Min Range: -5.3953

Max Act: 4.9891. Min Act: -0.1700

Data Index: 366685 (C4 (Web Text))

Max Activating Token Index: 626

Text #13

Max Range: 5.3953. Min Range: -5.3953

Max Act: 4.9578. Min Act: -0.1700

Data Index: 5512 (C4 (Web Text))

Max Activating Token Index: 395

Text #14

Max Range: 5.3953. Min Range: -5.3953

Max Act: 4.9578. Min Act: -0.1700

Data Index: 38992 (C4 (Web Text))

Max Activating Token Index: 806

Text #15

Max Range: 5.3953. Min Range: -5.3953

Max Act: 5.0203. Min Act: -0.1700

Data Index: 1113797 (C4 (Web Text))

Max Activating Token Index: 186

Text #16

Max Range: 5.3953. Min Range: -5.3953

Max Act: 4.9578. Min Act: -0.1700

Data Index: 524014 (C4 (Web Text))

Max Activating Token Index: 962

Text #17

Max Range: 5.3953. Min Range: -5.3953

Max Act: 4.9891. Min Act: -0.1700

Data Index: 55017 (C4 (Web Text))

Max Activating Token Index: 145

Text #18

Max Range: 5.3953. Min Range: -5.3953

Max Act: 4.9266. Min Act: -0.1700

Data Index: 482604 (C4 (Web Text))

Max Activating Token Index: 172

Text #19

Max Range: 5.3953. Min Range: -5.3953

Max Act: 4.9266. Min Act: -0.1700

Data Index: 1198919 (C4 (Web Text))

Max Activating Token Index: 1013