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

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

Neuron 136 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: 3.8072. Min Range: -3.8072

Max Act: 3.8072. Min Act: -0.1700

Data Index: 592199 (C4 (Web Text))

Max Activating Token Index: 185

Text #1

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.6820. Min Act: -0.1700

Data Index: 1317740 (C4 (Web Text))

Max Activating Token Index: 951

Text #2

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.6038. Min Act: -0.1700

Data Index: 624027 (C4 (Web Text))

Max Activating Token Index: 447

Text #3

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.5881. Min Act: -0.1700

Data Index: 836077 (C4 (Web Text))

Max Activating Token Index: 232

Text #4

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.6038. Min Act: -0.1700

Data Index: 1349165 (C4 (Web Text))

Max Activating Token Index: 830

Text #5

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.5255. Min Act: -0.1700

Data Index: 642865 (C4 (Web Text))

Max Activating Token Index: 208

Text #6

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.5411. Min Act: -0.1700

Data Index: 1106752 (C4 (Web Text))

Max Activating Token Index: 594

Text #7

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.5411. Min Act: -0.1700

Data Index: 113180 (C4 (Web Text))

Max Activating Token Index: 442

Text #8

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.5568. Min Act: -0.1700

Data Index: 102543 (C4 (Web Text))

Max Activating Token Index: 145

Text #9

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.5255. Min Act: -0.1700

Data Index: 408252 (C4 (Web Text))

Max Activating Token Index: 396

Text #10

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.4785. Min Act: -0.1700

Data Index: 88429 (C4 (Web Text))

Max Activating Token Index: 529

Text #11

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.4785. Min Act: -0.1700

Data Index: 1156541 (C4 (Web Text))

Max Activating Token Index: 473

Text #12

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.5255. Min Act: -0.1700

Data Index: 931696 (C4 (Web Text))

Max Activating Token Index: 584

Text #13

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.5098. Min Act: -0.1700

Data Index: 22963 (C4 (Web Text))

Max Activating Token Index: 686

Text #14

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.4785. Min Act: -0.1700

Data Index: 230284 (C4 (Web Text))

Max Activating Token Index: 40

Text #15

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.4941. Min Act: -0.1700

Data Index: 1018364 (C4 (Web Text))

Max Activating Token Index: 779

Text #16

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.4941. Min Act: -0.1700

Data Index: 1294095 (C4 (Web Text))

Max Activating Token Index: 207

Text #17

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.4628. Min Act: -0.1700

Data Index: 1033018 (C4 (Web Text))

Max Activating Token Index: 619

Text #18

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.4471. Min Act: -0.1700

Data Index: 1058228 (C4 (Web Text))

Max Activating Token Index: 969

Text #19

Max Range: 3.8072. Min Range: -3.8072

Max Act: 3.5098. Min Act: -0.1700

Data Index: 293633 (C4 (Web Text))

Max Activating Token Index: 571