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

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

Neuron 1031 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: 3.2963. Min Range: -3.2963

Max Act: 3.2963. Min Act: -0.1700

Data Index: 1315063 (C4 (Web Text))

Max Activating Token Index: 664

Text #1

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.2335. Min Act: -0.1700

Data Index: 713851 (C4 (Web Text))

Max Activating Token Index: 583

Text #2

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.2335. Min Act: -0.1700

Data Index: 600548 (C4 (Web Text))

Max Activating Token Index: 653

Text #3

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.1863. Min Act: -0.1700

Data Index: 511122 (C4 (Web Text))

Max Activating Token Index: 713

Text #4

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.1548. Min Act: -0.1700

Data Index: 419183 (C4 (Web Text))

Max Activating Token Index: 760

Text #5

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.1391. Min Act: -0.1700

Data Index: 1233379 (C4 (Web Text))

Max Activating Token Index: 272

Text #6

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.1391. Min Act: -0.1700

Data Index: 1316593 (C4 (Web Text))

Max Activating Token Index: 150

Text #7

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.1233. Min Act: -0.1700

Data Index: 677473 (C4 (Web Text))

Max Activating Token Index: 409

Text #8

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.1233. Min Act: -0.1700

Data Index: 234358 (C4 (Web Text))

Max Activating Token Index: 899

Text #9

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.1075. Min Act: -0.1700

Data Index: 1265212 (C4 (Web Text))

Max Activating Token Index: 593

Text #10

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.1075. Min Act: -0.1700

Data Index: 555949 (C4 (Web Text))

Max Activating Token Index: 133

Text #11

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.1233. Min Act: -0.1700

Data Index: 1233340 (C4 (Web Text))

Max Activating Token Index: 711

Text #12

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.0918. Min Act: -0.1700

Data Index: 43396 (C4 (Web Text))

Max Activating Token Index: 81

Text #13

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.0444. Min Act: -0.1700

Data Index: 153327 (C4 (Web Text))

Max Activating Token Index: 713

Text #14

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.0760. Min Act: -0.1700

Data Index: 1007659 (C4 (Web Text))

Max Activating Token Index: 852

Text #15

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.0760. Min Act: -0.1700

Data Index: 948705 (C4 (Web Text))

Max Activating Token Index: 434

Text #16

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.0602. Min Act: -0.1700

Data Index: 108175 (C4 (Web Text))

Max Activating Token Index: 861

Text #17

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.0444. Min Act: -0.1700

Data Index: 272957 (C4 (Web Text))

Max Activating Token Index: 725

Text #18

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.0918. Min Act: -0.1700

Data Index: 630424 (C4 (Web Text))

Max Activating Token Index: 963

Text #19

Max Range: 3.2963. Min Range: -3.2963

Max Act: 3.0444. Min Act: -0.1700

Data Index: 150199 (C4 (Web Text))

Max Activating Token Index: 438