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

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

Neuron 757 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: 4.4164. Min Range: -4.4164

Max Act: 4.4164. Min Act: -0.1700

Data Index: 224856 (C4 (Web Text))

Max Activating Token Index: 665

Text #1

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.2914. Min Act: -0.1700

Data Index: 570267 (C4 (Web Text))

Max Activating Token Index: 41

Text #2

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.2132. Min Act: -0.1700

Data Index: 438494 (C4 (Web Text))

Max Activating Token Index: 998

Text #3

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.1507. Min Act: -0.1700

Data Index: 1120462 (C4 (Web Text))

Max Activating Token Index: 966

Text #4

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.1820. Min Act: -0.1700

Data Index: 1005906 (C4 (Web Text))

Max Activating Token Index: 448

Text #5

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.1976. Min Act: -0.1700

Data Index: 1043505 (C4 (Web Text))

Max Activating Token Index: 1015

Text #6

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.1663. Min Act: -0.1700

Data Index: 867834 (C4 (Web Text))

Max Activating Token Index: 672

Text #7

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.1351. Min Act: -0.1700

Data Index: 20007 (C4 (Web Text))

Max Activating Token Index: 355

Text #8

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.1663. Min Act: -0.1700

Data Index: 903238 (C4 (Web Text))

Max Activating Token Index: 332

Text #9

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.1976. Min Act: -0.1700

Data Index: 622097 (C4 (Web Text))

Max Activating Token Index: 895

Text #10

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.1351. Min Act: -0.1700

Data Index: 1228203 (C4 (Web Text))

Max Activating Token Index: 519

Text #11

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.1507. Min Act: -0.1700

Data Index: 385100 (C4 (Web Text))

Max Activating Token Index: 635

Text #12

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.0726. Min Act: -0.1700

Data Index: 465955 (C4 (Web Text))

Max Activating Token Index: 72

Text #13

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.0882. Min Act: -0.1700

Data Index: 996122 (C4 (Web Text))

Max Activating Token Index: 469

Text #14

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.0726. Min Act: -0.1700

Data Index: 410463 (C4 (Web Text))

Max Activating Token Index: 14

Text #15

Max Range: 4.4164. Min Range: -4.4164

Max Act: 3.9319. Min Act: -0.1700

Data Index: 1593560 (Python Code)

Max Activating Token Index: 694

Text #16

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.0726. Min Act: -0.1699

Data Index: 1553891 (Python Code)

Max Activating Token Index: 773

Text #17

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.0257. Min Act: -0.1700

Data Index: 1232823 (C4 (Web Text))

Max Activating Token Index: 935

Text #18

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.0413. Min Act: -0.1700

Data Index: 191856 (C4 (Web Text))

Max Activating Token Index: 395

Text #19

Max Range: 4.4164. Min Range: -4.4164

Max Act: 4.0257. Min Act: -0.1700

Data Index: 597454 (C4 (Web Text))

Max Activating Token Index: 452