Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4490 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 3.5056. Min Range: -3.5056

Max Act: 3.5056. Min Act: -0.0000

Data Index: 767030 (C4 (Web Text))

Max Activating Token Index: 473

Text #1

Max Range: 3.5056. Min Range: -3.5056

Max Act: 2.3633. Min Act: -0.0000

Data Index: 1655040 (Python Code)

Max Activating Token Index: 498

Text #2

Max Range: 3.5056. Min Range: -3.5056

Max Act: 2.5066. Min Act: -0.0000

Data Index: 1527340 (Python Code)

Max Activating Token Index: 775

Text #3

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.9715. Min Act: -0.0000

Data Index: 1364016 (Python Code)

Max Activating Token Index: 493

Text #4

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.6991. Min Act: -0.0000

Data Index: 1090442 (C4 (Web Text))

Max Activating Token Index: 101

Text #5

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.8508. Min Act: -0.0000

Data Index: 1253853 (C4 (Web Text))

Max Activating Token Index: 236

Text #6

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.6670. Min Act: -0.0000

Data Index: 1242153 (C4 (Web Text))

Max Activating Token Index: 425

Text #7

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.7829. Min Act: -0.0000

Data Index: 1409266 (Python Code)

Max Activating Token Index: 986

Text #8

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.7781. Min Act: -0.0000

Data Index: 791643 (C4 (Web Text))

Max Activating Token Index: 767

Text #9

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.7611. Min Act: -0.0000

Data Index: 1190048 (C4 (Web Text))

Max Activating Token Index: 126

Text #10

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.8333. Min Act: -0.0000

Data Index: 550216 (C4 (Web Text))

Max Activating Token Index: 734

Text #11

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.6152. Min Act: -0.0000

Data Index: 1448523 (Python Code)

Max Activating Token Index: 852

Text #12

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.8309. Min Act: -0.0000

Data Index: 1084650 (C4 (Web Text))

Max Activating Token Index: 164

Text #13

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.4751. Min Act: -0.0000

Data Index: 1180079 (C4 (Web Text))

Max Activating Token Index: 210

Text #14

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.7141. Min Act: -0.0000

Data Index: 592486 (C4 (Web Text))

Max Activating Token Index: 961

Text #15

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.7617. Min Act: -0.0000

Data Index: 1363589 (Python Code)

Max Activating Token Index: 292

Text #16

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.6155. Min Act: -0.0000

Data Index: 215089 (C4 (Web Text))

Max Activating Token Index: 542

Text #17

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.7488. Min Act: -0.0000

Data Index: 1414836 (Python Code)

Max Activating Token Index: 696

Text #18

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.6131. Min Act: -0.0000

Data Index: 277728 (C4 (Web Text))

Max Activating Token Index: 723

Text #19

Max Range: 3.5056. Min Range: -3.5056

Max Act: 1.5931. Min Act: -0.0000

Data Index: 603614 (C4 (Web Text))

Max Activating Token Index: 508