Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 4 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.4644. Min Act: -0.1700

Data Index: 393998 (C4 (Web Text))

Max Activating Token Index: 517

Text #1

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.4330. Min Act: -0.1700

Data Index: 1458017 (Python Code)

Max Activating Token Index: 13

Text #2

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.4173. Min Act: -0.1700

Data Index: 426512 (C4 (Web Text))

Max Activating Token Index: 518

Text #3

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.4330. Min Act: -0.1700

Data Index: 1622858 (Python Code)

Max Activating Token Index: 616

Text #4

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.4016. Min Act: -0.1700

Data Index: 19693 (C4 (Web Text))

Max Activating Token Index: 403

Text #5

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.4016. Min Act: -0.1700

Data Index: 1503940 (Python Code)

Max Activating Token Index: 196

Text #6

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3546. Min Act: -0.1700

Data Index: 139155 (C4 (Web Text))

Max Activating Token Index: 225

Text #7

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3859. Min Act: -0.1700

Data Index: 1541182 (Python Code)

Max Activating Token Index: 137

Text #8

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3859. Min Act: -0.1700

Data Index: 367797 (C4 (Web Text))

Max Activating Token Index: 255

Text #9

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3546. Min Act: -0.1700

Data Index: 1435035 (Python Code)

Max Activating Token Index: 770

Text #10

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3703. Min Act: -0.1700

Data Index: 1408807 (Python Code)

Max Activating Token Index: 562

Text #11

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3703. Min Act: -0.1700

Data Index: 636318 (C4 (Web Text))

Max Activating Token Index: 22

Text #12

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3703. Min Act: -0.1700

Data Index: 538776 (C4 (Web Text))

Max Activating Token Index: 394

Text #13

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3546. Min Act: -0.1700

Data Index: 567827 (C4 (Web Text))

Max Activating Token Index: 625

Text #14

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3546. Min Act: -0.1700

Data Index: 1581058 (Python Code)

Max Activating Token Index: 891

Text #15

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3703. Min Act: -0.1700

Data Index: 110475 (C4 (Web Text))

Max Activating Token Index: 443

Text #16

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3232. Min Act: -0.1700

Data Index: 424220 (C4 (Web Text))

Max Activating Token Index: 413

Text #17

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3075. Min Act: -0.1700

Data Index: 1414076 (Python Code)

Max Activating Token Index: 266

Text #18

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3232. Min Act: -0.1700

Data Index: 774432 (C4 (Web Text))

Max Activating Token Index: 323

Text #19

Max Range: 3.4644. Min Range: -3.4644

Max Act: 3.3232. Min Act: -0.1700

Data Index: 695817 (C4 (Web Text))

Max Activating Token Index: 843