Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 261 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.6048. Min Act: -0.1700

Data Index: 680196 (C4 (Web Text))

Max Activating Token Index: 837

Text #1

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.4798. Min Act: -0.1700

Data Index: 1448986 (Python Code)

Max Activating Token Index: 1004

Text #2

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.3236. Min Act: -0.1700

Data Index: 695310 (C4 (Web Text))

Max Activating Token Index: 32

Text #3

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.3548. Min Act: -0.1700

Data Index: 510967 (C4 (Web Text))

Max Activating Token Index: 1004

Text #4

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.2923. Min Act: -0.1700

Data Index: 228761 (C4 (Web Text))

Max Activating Token Index: 902

Text #5

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.2611. Min Act: -0.1700

Data Index: 285979 (C4 (Web Text))

Max Activating Token Index: 634

Text #6

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.2298. Min Act: -0.1700

Data Index: 1078218 (C4 (Web Text))

Max Activating Token Index: 938

Text #7

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.3236. Min Act: -0.1700

Data Index: 400656 (C4 (Web Text))

Max Activating Token Index: 845

Text #8

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1986. Min Act: -0.1700

Data Index: 426190 (C4 (Web Text))

Max Activating Token Index: 834

Text #9

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1986. Min Act: -0.1700

Data Index: 675640 (C4 (Web Text))

Max Activating Token Index: 798

Text #10

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1986. Min Act: -0.1700

Data Index: 500773 (C4 (Web Text))

Max Activating Token Index: 500

Text #11

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1986. Min Act: -0.1700

Data Index: 1487730 (Python Code)

Max Activating Token Index: 174

Text #12

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1986. Min Act: -0.1700

Data Index: 790383 (C4 (Web Text))

Max Activating Token Index: 925

Text #13

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1673. Min Act: -0.1700

Data Index: 563738 (C4 (Web Text))

Max Activating Token Index: 853

Text #14

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1986. Min Act: -0.1700

Data Index: 47823 (C4 (Web Text))

Max Activating Token Index: 306

Text #15

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1986. Min Act: -0.1700

Data Index: 220183 (C4 (Web Text))

Max Activating Token Index: 11

Text #16

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1673. Min Act: -0.1700

Data Index: 1214938 (C4 (Web Text))

Max Activating Token Index: 346

Text #17

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1361. Min Act: -0.1700

Data Index: 714260 (C4 (Web Text))

Max Activating Token Index: 856

Text #18

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1361. Min Act: -0.1700

Data Index: 676477 (C4 (Web Text))

Max Activating Token Index: 72

Text #19

Max Range: 5.6048. Min Range: -5.6048

Max Act: 5.1673. Min Act: -0.1700

Data Index: 1626522 (Python Code)

Max Activating Token Index: 551