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

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

Neuron 944 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8664. Min Act: -0.1700

Data Index: 1026054 (C4 (Web Text))

Max Activating Token Index: 121

Text #1

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8507. Min Act: -0.1700

Data Index: 532706 (C4 (Web Text))

Max Activating Token Index: 77

Text #2

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8507. Min Act: -0.1700

Data Index: 1349826 (C4 (Web Text))

Max Activating Token Index: 181

Text #3

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8194. Min Act: -0.1700

Data Index: 409957 (C4 (Web Text))

Max Activating Token Index: 923

Text #4

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8351. Min Act: -0.1700

Data Index: 1109243 (C4 (Web Text))

Max Activating Token Index: 152

Text #5

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8507. Min Act: -0.1700

Data Index: 612422 (C4 (Web Text))

Max Activating Token Index: 419

Text #6

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8507. Min Act: -0.1700

Data Index: 968394 (C4 (Web Text))

Max Activating Token Index: 351

Text #7

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8194. Min Act: -0.1700

Data Index: 6049 (C4 (Web Text))

Max Activating Token Index: 74

Text #8

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8038. Min Act: -0.1700

Data Index: 1343077 (C4 (Web Text))

Max Activating Token Index: 890

Text #9

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8351. Min Act: -0.1700

Data Index: 93504 (C4 (Web Text))

Max Activating Token Index: 719

Text #10

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8194. Min Act: -0.1700

Data Index: 1211246 (C4 (Web Text))

Max Activating Token Index: 74

Text #11

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8038. Min Act: -0.1700

Data Index: 924738 (C4 (Web Text))

Max Activating Token Index: 780

Text #12

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8038. Min Act: -0.1700

Data Index: 329288 (C4 (Web Text))

Max Activating Token Index: 792

Text #13

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8194. Min Act: -0.1700

Data Index: 104623 (C4 (Web Text))

Max Activating Token Index: 890

Text #14

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8194. Min Act: -0.1700

Data Index: 703228 (C4 (Web Text))

Max Activating Token Index: 80

Text #15

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8194. Min Act: -0.1700

Data Index: 244512 (C4 (Web Text))

Max Activating Token Index: 807

Text #16

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8038. Min Act: -0.1700

Data Index: 1457895 (Python Code)

Max Activating Token Index: 32

Text #17

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.7882. Min Act: -0.1700

Data Index: 693134 (C4 (Web Text))

Max Activating Token Index: 897

Text #18

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8038. Min Act: -0.1700

Data Index: 649256 (C4 (Web Text))

Max Activating Token Index: 979

Text #19

Max Range: 3.8664. Min Range: -3.8664

Max Act: 3.8038. Min Act: -0.1700

Data Index: 366764 (C4 (Web Text))

Max Activating Token Index: 883