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

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

Neuron 1808 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.3474. Min Act: -0.1700

Data Index: 1332990 (C4 (Web Text))

Max Activating Token Index: 358

Text #1

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.2531. Min Act: -0.1700

Data Index: 414469 (C4 (Web Text))

Max Activating Token Index: 925

Text #2

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.2374. Min Act: -0.1700

Data Index: 549952 (C4 (Web Text))

Max Activating Token Index: 329

Text #3

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.2217. Min Act: -0.1700

Data Index: 664144 (C4 (Web Text))

Max Activating Token Index: 56

Text #4

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.2060. Min Act: -0.1700

Data Index: 642896 (C4 (Web Text))

Max Activating Token Index: 979

Text #5

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.2060. Min Act: -0.1700

Data Index: 84191 (C4 (Web Text))

Max Activating Token Index: 523

Text #6

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.2060. Min Act: -0.1700

Data Index: 1141717 (C4 (Web Text))

Max Activating Token Index: 1022

Text #7

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1902. Min Act: -0.1700

Data Index: 510421 (C4 (Web Text))

Max Activating Token Index: 578

Text #8

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1745. Min Act: -0.1700

Data Index: 1010565 (C4 (Web Text))

Max Activating Token Index: 174

Text #9

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1587. Min Act: -0.1700

Data Index: 1199930 (C4 (Web Text))

Max Activating Token Index: 866

Text #10

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1430. Min Act: -0.1700

Data Index: 1540524 (Python Code)

Max Activating Token Index: 913

Text #11

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1430. Min Act: -0.1700

Data Index: 1313307 (C4 (Web Text))

Max Activating Token Index: 797

Text #12

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1272. Min Act: -0.1700

Data Index: 218974 (C4 (Web Text))

Max Activating Token Index: 419

Text #13

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1272. Min Act: -0.1700

Data Index: 725172 (C4 (Web Text))

Max Activating Token Index: 411

Text #14

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1272. Min Act: -0.1700

Data Index: 1537342 (Python Code)

Max Activating Token Index: 688

Text #15

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1272. Min Act: -0.1700

Data Index: 137270 (C4 (Web Text))

Max Activating Token Index: 133

Text #16

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.1115. Min Act: -0.1700

Data Index: 410821 (C4 (Web Text))

Max Activating Token Index: 580

Text #17

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.0957. Min Act: -0.1700

Data Index: 1119512 (C4 (Web Text))

Max Activating Token Index: 736

Text #18

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.0957. Min Act: -0.1700

Data Index: 1032468 (C4 (Web Text))

Max Activating Token Index: 764

Text #19

Max Range: 3.3474. Min Range: -3.3474

Max Act: 3.0957. Min Act: -0.1700

Data Index: 236499 (C4 (Web Text))

Max Activating Token Index: 424