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

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

Neuron 1797 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.8678. Min Act: -0.1700

Data Index: 1079408 (C4 (Web Text))

Max Activating Token Index: 275

Text #1

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.8365. Min Act: -0.1700

Data Index: 801099 (C4 (Web Text))

Max Activating Token Index: 591

Text #2

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.8052. Min Act: -0.1700

Data Index: 408760 (C4 (Web Text))

Max Activating Token Index: 851

Text #3

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.7427. Min Act: -0.1700

Data Index: 959478 (C4 (Web Text))

Max Activating Token Index: 889

Text #4

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.6488. Min Act: -0.1700

Data Index: 105470 (C4 (Web Text))

Max Activating Token Index: 798

Text #5

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.6488. Min Act: -0.1700

Data Index: 1136148 (C4 (Web Text))

Max Activating Token Index: 288

Text #6

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.6488. Min Act: -0.1700

Data Index: 500573 (C4 (Web Text))

Max Activating Token Index: 658

Text #7

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.6488. Min Act: -0.1700

Data Index: 201235 (C4 (Web Text))

Max Activating Token Index: 910

Text #8

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5861. Min Act: -0.1700

Data Index: 1510489 (Python Code)

Max Activating Token Index: 814

Text #9

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5861. Min Act: -0.1700

Data Index: 1153631 (C4 (Web Text))

Max Activating Token Index: 205

Text #10

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5861. Min Act: -0.1700

Data Index: 1238506 (C4 (Web Text))

Max Activating Token Index: 261

Text #11

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5861. Min Act: -0.1700

Data Index: 1145585 (C4 (Web Text))

Max Activating Token Index: 608

Text #12

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5548. Min Act: -0.1700

Data Index: 613668 (C4 (Web Text))

Max Activating Token Index: 206

Text #13

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5235. Min Act: -0.1700

Data Index: 873809 (C4 (Web Text))

Max Activating Token Index: 529

Text #14

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5861. Min Act: -0.1700

Data Index: 392719 (C4 (Web Text))

Max Activating Token Index: 34

Text #15

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5235. Min Act: -0.1700

Data Index: 127457 (C4 (Web Text))

Max Activating Token Index: 644

Text #16

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5235. Min Act: -0.1700

Data Index: 1075191 (C4 (Web Text))

Max Activating Token Index: 190

Text #17

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5235. Min Act: -0.1700

Data Index: 145306 (C4 (Web Text))

Max Activating Token Index: 299

Text #18

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.5235. Min Act: -0.1700

Data Index: 992321 (C4 (Web Text))

Max Activating Token Index: 838

Text #19

Max Range: 3.8678. Min Range: -3.8678

Max Act: 3.4922. Min Act: -0.1700

Data Index: 375849 (C4 (Web Text))

Max Activating Token Index: 311