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

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

Neuron 579 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: 5.0879. Min Range: -5.0879

Max Act: 5.0879. Min Act: -0.1700

Data Index: 816195 (C4 (Web Text))

Max Activating Token Index: 284

Text #1

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.9004. Min Act: -0.1700

Data Index: 714223 (C4 (Web Text))

Max Activating Token Index: 834

Text #2

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.9316. Min Act: -0.1700

Data Index: 196649 (C4 (Web Text))

Max Activating Token Index: 33

Text #3

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.8691. Min Act: -0.1700

Data Index: 1205025 (C4 (Web Text))

Max Activating Token Index: 832

Text #4

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.8066. Min Act: -0.1700

Data Index: 201382 (C4 (Web Text))

Max Activating Token Index: 85

Text #5

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7597. Min Act: -0.1700

Data Index: 84587 (C4 (Web Text))

Max Activating Token Index: 969

Text #6

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7910. Min Act: -0.1700

Data Index: 1251341 (C4 (Web Text))

Max Activating Token Index: 910

Text #7

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7597. Min Act: -0.1700

Data Index: 466192 (C4 (Web Text))

Max Activating Token Index: 666

Text #8

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7441. Min Act: -0.1700

Data Index: 503335 (C4 (Web Text))

Max Activating Token Index: 38

Text #9

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7129. Min Act: -0.1700

Data Index: 456496 (C4 (Web Text))

Max Activating Token Index: 660

Text #10

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7441. Min Act: -0.1700

Data Index: 620726 (C4 (Web Text))

Max Activating Token Index: 425

Text #11

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7285. Min Act: -0.1700

Data Index: 398974 (C4 (Web Text))

Max Activating Token Index: 870

Text #12

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.6816. Min Act: -0.1700

Data Index: 1146753 (C4 (Web Text))

Max Activating Token Index: 489

Text #13

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7285. Min Act: -0.1700

Data Index: 104909 (C4 (Web Text))

Max Activating Token Index: 464

Text #14

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7285. Min Act: -0.1700

Data Index: 891726 (C4 (Web Text))

Max Activating Token Index: 97

Text #15

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.6816. Min Act: -0.1700

Data Index: 1337241 (C4 (Web Text))

Max Activating Token Index: 837

Text #16

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.6972. Min Act: -0.1700

Data Index: 851720 (C4 (Web Text))

Max Activating Token Index: 849

Text #17

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7129. Min Act: -0.1700

Data Index: 89539 (C4 (Web Text))

Max Activating Token Index: 652

Text #18

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.6504. Min Act: -0.1700

Data Index: 137474 (C4 (Web Text))

Max Activating Token Index: 404

Text #19

Max Range: 5.0879. Min Range: -5.0879

Max Act: 4.7129. Min Act: -0.1700

Data Index: 769322 (C4 (Web Text))

Max Activating Token Index: 552