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

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

Neuron 2041 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: 5.2957. Min Range: -5.2957

Max Act: 5.2957. Min Act: -0.1700

Data Index: 1505340 (Python Code)

Max Activating Token Index: 461

Text #1

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.2644. Min Act: -0.1700

Data Index: 1582558 (Python Code)

Max Activating Token Index: 778

Text #2

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.1863. Min Act: -0.1700

Data Index: 1399076 (Python Code)

Max Activating Token Index: 747

Text #3

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.1394. Min Act: -0.1700

Data Index: 1474365 (Python Code)

Max Activating Token Index: 753

Text #4

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.1394. Min Act: -0.1700

Data Index: 1466209 (Python Code)

Max Activating Token Index: 788

Text #5

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.0613. Min Act: -0.1700

Data Index: 1513680 (Python Code)

Max Activating Token Index: 1012

Text #6

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.1238. Min Act: -0.1700

Data Index: 1454212 (Python Code)

Max Activating Token Index: 989

Text #7

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.0457. Min Act: -0.1700

Data Index: 1490714 (Python Code)

Max Activating Token Index: 496

Text #8

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.0301. Min Act: -0.1700

Data Index: 1595727 (Python Code)

Max Activating Token Index: 203

Text #9

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.0144. Min Act: -0.1700

Data Index: 1544507 (Python Code)

Max Activating Token Index: 965

Text #10

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.0926. Min Act: -0.1700

Data Index: 1519498 (Python Code)

Max Activating Token Index: 46

Text #11

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.0613. Min Act: -0.1700

Data Index: 1441407 (Python Code)

Max Activating Token Index: 148

Text #12

Max Range: 5.2957. Min Range: -5.2957

Max Act: 5.0144. Min Act: -0.1700

Data Index: 628036 (C4 (Web Text))

Max Activating Token Index: 370

Text #13

Max Range: 5.2957. Min Range: -5.2957

Max Act: 4.9676. Min Act: -0.1700

Data Index: 742552 (C4 (Web Text))

Max Activating Token Index: 1015

Text #14

Max Range: 5.2957. Min Range: -5.2957

Max Act: 4.8894. Min Act: -0.1700

Data Index: 1492646 (Python Code)

Max Activating Token Index: 396

Text #15

Max Range: 5.2957. Min Range: -5.2957

Max Act: 4.9676. Min Act: -0.1700

Data Index: 1534096 (Python Code)

Max Activating Token Index: 993

Text #16

Max Range: 5.2957. Min Range: -5.2957

Max Act: 4.9519. Min Act: -0.1700

Data Index: 1572138 (Python Code)

Max Activating Token Index: 772

Text #17

Max Range: 5.2957. Min Range: -5.2957

Max Act: 4.8738. Min Act: -0.1700

Data Index: 1572817 (Python Code)

Max Activating Token Index: 886

Text #18

Max Range: 5.2957. Min Range: -5.2957

Max Act: 4.8738. Min Act: -0.1700

Data Index: 1483219 (Python Code)

Max Activating Token Index: 440

Text #19

Max Range: 5.2957. Min Range: -5.2957

Max Act: 4.8582. Min Act: -0.1700

Data Index: 1595732 (Python Code)

Max Activating Token Index: 663