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

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

Neuron 511 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.6110. Min Act: -0.1700

Data Index: 480631 (C4 (Web Text))

Max Activating Token Index: 632

Text #1

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.5485. Min Act: -0.1700

Data Index: 97770 (C4 (Web Text))

Max Activating Token Index: 948

Text #2

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.4860. Min Act: -0.1700

Data Index: 1513830 (Python Code)

Max Activating Token Index: 770

Text #3

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.5173. Min Act: -0.1699

Data Index: 125199 (C4 (Web Text))

Max Activating Token Index: 851

Text #4

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.4860. Min Act: -0.1700

Data Index: 1360398 (Python Code)

Max Activating Token Index: 250

Text #5

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.4548. Min Act: -0.1700

Data Index: 651277 (C4 (Web Text))

Max Activating Token Index: 685

Text #6

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.4860. Min Act: -0.1700

Data Index: 1438583 (Python Code)

Max Activating Token Index: 801

Text #7

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.3923. Min Act: -0.1700

Data Index: 113116 (C4 (Web Text))

Max Activating Token Index: 513

Text #8

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.3610. Min Act: -0.1700

Data Index: 1546818 (Python Code)

Max Activating Token Index: 805

Text #9

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.2985. Min Act: -0.1700

Data Index: 1452383 (Python Code)

Max Activating Token Index: 1022

Text #10

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.3610. Min Act: -0.1700

Data Index: 166289 (C4 (Web Text))

Max Activating Token Index: 372

Text #11

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.3298. Min Act: -0.1700

Data Index: 745047 (C4 (Web Text))

Max Activating Token Index: 879

Text #12

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.2985. Min Act: -0.1700

Data Index: 1437400 (Python Code)

Max Activating Token Index: 480

Text #13

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.2673. Min Act: -0.1700

Data Index: 1561728 (Python Code)

Max Activating Token Index: 1004

Text #14

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.2985. Min Act: -0.1700

Data Index: 1131493 (C4 (Web Text))

Max Activating Token Index: 450

Text #15

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.2673. Min Act: -0.1700

Data Index: 323186 (C4 (Web Text))

Max Activating Token Index: 872

Text #16

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.2360. Min Act: -0.1700

Data Index: 1551480 (Python Code)

Max Activating Token Index: 598

Text #17

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.2360. Min Act: -0.1700

Data Index: 1021027 (C4 (Web Text))

Max Activating Token Index: 319

Text #18

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.2048. Min Act: -0.1700

Data Index: 493505 (C4 (Web Text))

Max Activating Token Index: 525

Text #19

Max Range: 5.6110. Min Range: -5.6110

Max Act: 5.2360. Min Act: -0.1700

Data Index: 1390062 (Python Code)

Max Activating Token Index: 276