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

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

Neuron 1097 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.9579. Min Act: -0.0000

Data Index: 1109977 (C4 (Web Text))

Max Activating Token Index: 571

Text #1

Max Range: 2.1648. Min Range: -2.1648

Max Act: 2.1648. Min Act: -0.0000

Data Index: 217701 (C4 (Web Text))

Max Activating Token Index: 134

Text #2

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.9545. Min Act: -0.0001

Data Index: 52702 (C4 (Web Text))

Max Activating Token Index: 993

Text #3

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.7232. Min Act: -0.0000

Data Index: 139749 (C4 (Web Text))

Max Activating Token Index: 189

Text #4

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.9460. Min Act: -0.0000

Data Index: 105078 (C4 (Web Text))

Max Activating Token Index: 23

Text #5

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.7893. Min Act: -0.0000

Data Index: 84544 (C4 (Web Text))

Max Activating Token Index: 24

Text #6

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.6318. Min Act: -0.0000

Data Index: 1513006 (Python Code)

Max Activating Token Index: 398

Text #7

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.8445. Min Act: -0.0000

Data Index: 587201 (C4 (Web Text))

Max Activating Token Index: 326

Text #8

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.8485. Min Act: -0.0000

Data Index: 1591740 (Python Code)

Max Activating Token Index: 668

Text #9

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.6524. Min Act: -0.0001

Data Index: 902136 (C4 (Web Text))

Max Activating Token Index: 15

Text #10

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.5587. Min Act: -0.0000

Data Index: 635290 (C4 (Web Text))

Max Activating Token Index: 87

Text #11

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.5552. Min Act: -0.0000

Data Index: 281671 (C4 (Web Text))

Max Activating Token Index: 364

Text #12

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.6170. Min Act: -0.0000

Data Index: 1481216 (Python Code)

Max Activating Token Index: 281

Text #13

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.6106. Min Act: -0.0001

Data Index: 1630153 (Python Code)

Max Activating Token Index: 160

Text #14

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.5850. Min Act: -0.0000

Data Index: 1396431 (Python Code)

Max Activating Token Index: 143

Text #15

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.5969. Min Act: -0.0000

Data Index: 852749 (C4 (Web Text))

Max Activating Token Index: 604

Text #16

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.3560. Min Act: -0.0000

Data Index: 72134 (C4 (Web Text))

Max Activating Token Index: 246

Text #17

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.4880. Min Act: -0.0000

Data Index: 1639183 (Python Code)

Max Activating Token Index: 23

Text #18

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.5674. Min Act: -0.0000

Data Index: 1515663 (Python Code)

Max Activating Token Index: 128

Text #19

Max Range: 2.1648. Min Range: -2.1648

Max Act: 1.5472. Min Act: -0.0000

Data Index: 1258771 (C4 (Web Text))

Max Activating Token Index: 130