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

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

Neuron 1686 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.2493. Min Act: -0.1700

Data Index: 955978 (C4 (Web Text))

Max Activating Token Index: 204

Text #1

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.2493. Min Act: -0.1700

Data Index: 1084975 (C4 (Web Text))

Max Activating Token Index: 747

Text #2

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.2179. Min Act: -0.1700

Data Index: 210038 (C4 (Web Text))

Max Activating Token Index: 915

Text #3

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.2179. Min Act: -0.1700

Data Index: 195666 (C4 (Web Text))

Max Activating Token Index: 864

Text #4

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.1392. Min Act: -0.1700

Data Index: 925026 (C4 (Web Text))

Max Activating Token Index: 398

Text #5

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.1549. Min Act: -0.1700

Data Index: 1288166 (C4 (Web Text))

Max Activating Token Index: 252

Text #6

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.1234. Min Act: -0.1700

Data Index: 1626179 (Python Code)

Max Activating Token Index: 935

Text #7

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.1392. Min Act: -0.1700

Data Index: 952092 (C4 (Web Text))

Max Activating Token Index: 844

Text #8

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.1077. Min Act: -0.1700

Data Index: 910850 (C4 (Web Text))

Max Activating Token Index: 980

Text #9

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.1234. Min Act: -0.1700

Data Index: 805913 (C4 (Web Text))

Max Activating Token Index: 71

Text #10

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.1077. Min Act: -0.1700

Data Index: 263681 (C4 (Web Text))

Max Activating Token Index: 890

Text #11

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.0919. Min Act: -0.1700

Data Index: 868669 (C4 (Web Text))

Max Activating Token Index: 652

Text #12

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.0919. Min Act: -0.1700

Data Index: 292280 (C4 (Web Text))

Max Activating Token Index: 167

Text #13

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.0761. Min Act: -0.1700

Data Index: 550183 (C4 (Web Text))

Max Activating Token Index: 824

Text #14

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.0919. Min Act: -0.1700

Data Index: 640040 (C4 (Web Text))

Max Activating Token Index: 712

Text #15

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.1077. Min Act: -0.1700

Data Index: 42286 (C4 (Web Text))

Max Activating Token Index: 471

Text #16

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.0919. Min Act: -0.1700

Data Index: 318926 (C4 (Web Text))

Max Activating Token Index: 531

Text #17

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.0761. Min Act: -0.1700

Data Index: 635376 (C4 (Web Text))

Max Activating Token Index: 75

Text #18

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.0603. Min Act: -0.1700

Data Index: 1538560 (Python Code)

Max Activating Token Index: 758

Text #19

Max Range: 3.2493. Min Range: -3.2493

Max Act: 3.0761. Min Act: -0.1700

Data Index: 1390460 (Python Code)

Max Activating Token Index: 827