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

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

Neuron 1195 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: 7.3403. Min Range: -7.3403

Max Act: 7.3403. Min Act: -0.1700

Data Index: 1542178 (Python Code)

Max Activating Token Index: 501

Text #1

Max Range: 7.3403. Min Range: -7.3403

Max Act: 7.3403. Min Act: -0.1700

Data Index: 945536 (C4 (Web Text))

Max Activating Token Index: 629

Text #2

Max Range: 7.3403. Min Range: -7.3403

Max Act: 7.1528. Min Act: -0.1700

Data Index: 1543924 (Python Code)

Max Activating Token Index: 481

Text #3

Max Range: 7.3403. Min Range: -7.3403

Max Act: 7.0591. Min Act: -0.1700

Data Index: 1550873 (Python Code)

Max Activating Token Index: 136

Text #4

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.9966. Min Act: -0.1700

Data Index: 1654014 (Python Code)

Max Activating Token Index: 139

Text #5

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.7778. Min Act: -0.1700

Data Index: 1376797 (Python Code)

Max Activating Token Index: 365

Text #6

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.7778. Min Act: -0.1700

Data Index: 1644584 (Python Code)

Max Activating Token Index: 341

Text #7

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.7466. Min Act: -0.1700

Data Index: 1552447 (Python Code)

Max Activating Token Index: 434

Text #8

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.7778. Min Act: -0.1700

Data Index: 1609642 (Python Code)

Max Activating Token Index: 723

Text #9

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.6841. Min Act: -0.1700

Data Index: 298898 (C4 (Web Text))

Max Activating Token Index: 609

Text #10

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.7153. Min Act: -0.1700

Data Index: 910950 (C4 (Web Text))

Max Activating Token Index: 536

Text #11

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.5591. Min Act: -0.1700

Data Index: 1642258 (Python Code)

Max Activating Token Index: 298

Text #12

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.6528. Min Act: -0.1700

Data Index: 176083 (C4 (Web Text))

Max Activating Token Index: 958

Text #13

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.7153. Min Act: -0.1699

Data Index: 694059 (C4 (Web Text))

Max Activating Token Index: 191

Text #14

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.5903. Min Act: -0.1700

Data Index: 167822 (C4 (Web Text))

Max Activating Token Index: 67

Text #15

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.6216. Min Act: -0.1700

Data Index: 1634701 (Python Code)

Max Activating Token Index: 330

Text #16

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.6216. Min Act: -0.1700

Data Index: 1530228 (Python Code)

Max Activating Token Index: 856

Text #17

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.6216. Min Act: -0.1699

Data Index: 1453320 (Python Code)

Max Activating Token Index: 97

Text #18

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.6841. Min Act: -0.1700

Data Index: 831370 (C4 (Web Text))

Max Activating Token Index: 228

Text #19

Max Range: 7.3403. Min Range: -7.3403

Max Act: 6.6216. Min Act: -0.1700

Data Index: 1211866 (C4 (Web Text))

Max Activating Token Index: 739