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

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

Neuron 651 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.8983. Min Act: -0.1700

Data Index: 550228 (C4 (Web Text))

Max Activating Token Index: 475

Text #1

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.8358. Min Act: -0.1700

Data Index: 166927 (C4 (Web Text))

Max Activating Token Index: 380

Text #2

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.7889. Min Act: -0.1700

Data Index: 521522 (C4 (Web Text))

Max Activating Token Index: 1005

Text #3

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.7107. Min Act: -0.1700

Data Index: 1165473 (C4 (Web Text))

Max Activating Token Index: 483

Text #4

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6794. Min Act: -0.1700

Data Index: 954591 (C4 (Web Text))

Max Activating Token Index: 676

Text #5

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.7263. Min Act: -0.1700

Data Index: 499649 (C4 (Web Text))

Max Activating Token Index: 970

Text #6

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6637. Min Act: -0.1700

Data Index: 1214778 (C4 (Web Text))

Max Activating Token Index: 209

Text #7

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6794. Min Act: -0.1700

Data Index: 12198 (C4 (Web Text))

Max Activating Token Index: 978

Text #8

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6794. Min Act: -0.1700

Data Index: 440432 (C4 (Web Text))

Max Activating Token Index: 269

Text #9

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6637. Min Act: -0.1700

Data Index: 1285465 (C4 (Web Text))

Max Activating Token Index: 792

Text #10

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.7107. Min Act: -0.1700

Data Index: 637401 (C4 (Web Text))

Max Activating Token Index: 404

Text #11

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6794. Min Act: -0.1700

Data Index: 1134773 (C4 (Web Text))

Max Activating Token Index: 240

Text #12

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6637. Min Act: -0.1700

Data Index: 1119641 (C4 (Web Text))

Max Activating Token Index: 980

Text #13

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6481. Min Act: -0.1700

Data Index: 373953 (C4 (Web Text))

Max Activating Token Index: 550

Text #14

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6637. Min Act: -0.1700

Data Index: 489315 (C4 (Web Text))

Max Activating Token Index: 730

Text #15

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6950. Min Act: -0.1700

Data Index: 1169830 (C4 (Web Text))

Max Activating Token Index: 169

Text #16

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6324. Min Act: -0.1700

Data Index: 1118213 (C4 (Web Text))

Max Activating Token Index: 15

Text #17

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6167. Min Act: -0.1700

Data Index: 766522 (C4 (Web Text))

Max Activating Token Index: 575

Text #18

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.5385. Min Act: -0.1700

Data Index: 1097063 (C4 (Web Text))

Max Activating Token Index: 659

Text #19

Max Range: 3.8983. Min Range: -3.8983

Max Act: 3.6324. Min Act: -0.1700

Data Index: 267899 (C4 (Web Text))

Max Activating Token Index: 733