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

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

Neuron 1580 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: 5.2769. Min Range: -5.2769

Max Act: 5.2769. Min Act: -0.1700

Data Index: 883542 (C4 (Web Text))

Max Activating Token Index: 51

Text #1

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.2457. Min Act: -0.1700

Data Index: 323240 (C4 (Web Text))

Max Activating Token Index: 590

Text #2

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.1832. Min Act: -0.1700

Data Index: 1319565 (C4 (Web Text))

Max Activating Token Index: 961

Text #3

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.2144. Min Act: -0.1700

Data Index: 845142 (C4 (Web Text))

Max Activating Token Index: 82

Text #4

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.1832. Min Act: -0.1700

Data Index: 467755 (C4 (Web Text))

Max Activating Token Index: 612

Text #5

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.1519. Min Act: -0.1700

Data Index: 847211 (C4 (Web Text))

Max Activating Token Index: 581

Text #6

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.0894. Min Act: -0.1700

Data Index: 855582 (C4 (Web Text))

Max Activating Token Index: 630

Text #7

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.1207. Min Act: -0.1700

Data Index: 903664 (C4 (Web Text))

Max Activating Token Index: 938

Text #8

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.0894. Min Act: -0.1700

Data Index: 194853 (C4 (Web Text))

Max Activating Token Index: 55

Text #9

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.0582. Min Act: -0.1700

Data Index: 900610 (C4 (Web Text))

Max Activating Token Index: 622

Text #10

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.0894. Min Act: -0.1700

Data Index: 791757 (C4 (Web Text))

Max Activating Token Index: 940

Text #11

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.0269. Min Act: -0.1700

Data Index: 982229 (C4 (Web Text))

Max Activating Token Index: 933

Text #12

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.0269. Min Act: -0.1700

Data Index: 1344116 (C4 (Web Text))

Max Activating Token Index: 749

Text #13

Max Range: 5.2769. Min Range: -5.2769

Max Act: 5.0582. Min Act: -0.1700

Data Index: 618706 (C4 (Web Text))

Max Activating Token Index: 983

Text #14

Max Range: 5.2769. Min Range: -5.2769

Max Act: 4.9644. Min Act: -0.1700

Data Index: 1270442 (C4 (Web Text))

Max Activating Token Index: 820

Text #15

Max Range: 5.2769. Min Range: -5.2769

Max Act: 4.9957. Min Act: -0.1700

Data Index: 1198380 (C4 (Web Text))

Max Activating Token Index: 499

Text #16

Max Range: 5.2769. Min Range: -5.2769

Max Act: 4.9644. Min Act: -0.1700

Data Index: 430535 (C4 (Web Text))

Max Activating Token Index: 460

Text #17

Max Range: 5.2769. Min Range: -5.2769

Max Act: 4.9957. Min Act: -0.1700

Data Index: 213385 (C4 (Web Text))

Max Activating Token Index: 174

Text #18

Max Range: 5.2769. Min Range: -5.2769

Max Act: 4.9644. Min Act: -0.1700

Data Index: 1021062 (C4 (Web Text))

Max Activating Token Index: 12

Text #19

Max Range: 5.2769. Min Range: -5.2769

Max Act: 4.9957. Min Act: -0.1700

Data Index: 494398 (C4 (Web Text))

Max Activating Token Index: 336