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

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

Neuron 1910 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.6084. Min Act: -0.1699

Data Index: 235385 (C4 (Web Text))

Max Activating Token Index: 156

Text #1

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.5771. Min Act: -0.1700

Data Index: 1230081 (C4 (Web Text))

Max Activating Token Index: 66

Text #2

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4831. Min Act: -0.1700

Data Index: 1188902 (C4 (Web Text))

Max Activating Token Index: 71

Text #3

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.5144. Min Act: -0.1700

Data Index: 988457 (C4 (Web Text))

Max Activating Token Index: 90

Text #4

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4831. Min Act: -0.1700

Data Index: 517924 (C4 (Web Text))

Max Activating Token Index: 276

Text #5

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4204. Min Act: -0.1700

Data Index: 1243126 (C4 (Web Text))

Max Activating Token Index: 129

Text #6

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4517. Min Act: -0.1700

Data Index: 174982 (C4 (Web Text))

Max Activating Token Index: 347

Text #7

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4517. Min Act: -0.1700

Data Index: 10700 (C4 (Web Text))

Max Activating Token Index: 194

Text #8

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4517. Min Act: -0.1700

Data Index: 756117 (C4 (Web Text))

Max Activating Token Index: 117

Text #9

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4517. Min Act: -0.1700

Data Index: 1286165 (C4 (Web Text))

Max Activating Token Index: 172

Text #10

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4831. Min Act: -0.1700

Data Index: 696282 (C4 (Web Text))

Max Activating Token Index: 55

Text #11

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4204. Min Act: -0.1700

Data Index: 1189501 (C4 (Web Text))

Max Activating Token Index: 911

Text #12

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4047. Min Act: -0.1700

Data Index: 1353459 (C4 (Web Text))

Max Activating Token Index: 610

Text #13

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.4047. Min Act: -0.1700

Data Index: 819143 (C4 (Web Text))

Max Activating Token Index: 720

Text #14

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.3890. Min Act: -0.1700

Data Index: 508812 (C4 (Web Text))

Max Activating Token Index: 143

Text #15

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.3733. Min Act: -0.1700

Data Index: 972984 (C4 (Web Text))

Max Activating Token Index: 35

Text #16

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.3733. Min Act: -0.1700

Data Index: 996297 (C4 (Web Text))

Max Activating Token Index: 194

Text #17

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.3890. Min Act: -0.1700

Data Index: 807250 (C4 (Web Text))

Max Activating Token Index: 109

Text #18

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.3890. Min Act: -0.1700

Data Index: 193169 (C4 (Web Text))

Max Activating Token Index: 73

Text #19

Max Range: 3.6084. Min Range: -3.6084

Max Act: 3.3890. Min Act: -0.1699

Data Index: 7903 (C4 (Web Text))

Max Activating Token Index: 419