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

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

Neuron 1655 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: 7.0584. Min Range: -7.0584

Max Act: 7.0584. Min Act: -0.1700

Data Index: 27046 (C4 (Web Text))

Max Activating Token Index: 659

Text #1

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.8396. Min Act: -0.1700

Data Index: 204964 (C4 (Web Text))

Max Activating Token Index: 699

Text #2

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.5896. Min Act: -0.1700

Data Index: 377544 (C4 (Web Text))

Max Activating Token Index: 350

Text #3

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.6209. Min Act: -0.1700

Data Index: 1007741 (C4 (Web Text))

Max Activating Token Index: 303

Text #4

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.6521. Min Act: -0.1700

Data Index: 495529 (C4 (Web Text))

Max Activating Token Index: 857

Text #5

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4959. Min Act: -0.1700

Data Index: 1034955 (C4 (Web Text))

Max Activating Token Index: 407

Text #6

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4959. Min Act: -0.1700

Data Index: 439147 (C4 (Web Text))

Max Activating Token Index: 588

Text #7

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.5584. Min Act: -0.1699

Data Index: 1149987 (C4 (Web Text))

Max Activating Token Index: 241

Text #8

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.5584. Min Act: -0.1700

Data Index: 253901 (C4 (Web Text))

Max Activating Token Index: 442

Text #9

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4959. Min Act: -0.1700

Data Index: 323005 (C4 (Web Text))

Max Activating Token Index: 656

Text #10

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4646. Min Act: -0.1700

Data Index: 1047306 (C4 (Web Text))

Max Activating Token Index: 511

Text #11

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4959. Min Act: -0.1700

Data Index: 838111 (C4 (Web Text))

Max Activating Token Index: 167

Text #12

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4959. Min Act: -0.1700

Data Index: 1155126 (C4 (Web Text))

Max Activating Token Index: 1003

Text #13

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4646. Min Act: -0.1700

Data Index: 951218 (C4 (Web Text))

Max Activating Token Index: 970

Text #14

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4021. Min Act: -0.1700

Data Index: 24589 (C4 (Web Text))

Max Activating Token Index: 173

Text #15

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4334. Min Act: -0.1700

Data Index: 1194209 (C4 (Web Text))

Max Activating Token Index: 721

Text #16

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.3084. Min Act: -0.1700

Data Index: 51987 (C4 (Web Text))

Max Activating Token Index: 190

Text #17

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.3709. Min Act: -0.1700

Data Index: 360261 (C4 (Web Text))

Max Activating Token Index: 276

Text #18

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.4334. Min Act: -0.1700

Data Index: 680925 (C4 (Web Text))

Max Activating Token Index: 101

Text #19

Max Range: 7.0584. Min Range: -7.0584

Max Act: 6.3709. Min Act: -0.1700

Data Index: 718073 (C4 (Web Text))

Max Activating Token Index: 815