Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2584 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.5986. Min Act: -0.0000

Data Index: 976791 (C4 (Web Text))

Max Activating Token Index: 1004

Text #1

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.4540. Min Act: -0.0000

Data Index: 848169 (C4 (Web Text))

Max Activating Token Index: 809

Text #2

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.3018. Min Act: -0.0000

Data Index: 297224 (C4 (Web Text))

Max Activating Token Index: 245

Text #3

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.1594. Min Act: -0.0000

Data Index: 1630702 (Python Code)

Max Activating Token Index: 234

Text #4

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.1147. Min Act: -0.0000

Data Index: 220576 (C4 (Web Text))

Max Activating Token Index: 862

Text #5

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.0607. Min Act: -0.0000

Data Index: 515564 (C4 (Web Text))

Max Activating Token Index: 644

Text #6

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.1351. Min Act: -0.0000

Data Index: 353650 (C4 (Web Text))

Max Activating Token Index: 120

Text #7

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.0625. Min Act: -0.0000

Data Index: 860769 (C4 (Web Text))

Max Activating Token Index: 250

Text #8

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.0410. Min Act: -0.0000

Data Index: 1163558 (C4 (Web Text))

Max Activating Token Index: 535

Text #9

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.0499. Min Act: -0.0000

Data Index: 300184 (C4 (Web Text))

Max Activating Token Index: 807

Text #10

Max Range: 1.5986. Min Range: -1.5986

Max Act: 0.9782. Min Act: -0.0000

Data Index: 960534 (C4 (Web Text))

Max Activating Token Index: 249

Text #11

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.0020. Min Act: -0.0000

Data Index: 369720 (C4 (Web Text))

Max Activating Token Index: 448

Text #12

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.0312. Min Act: -0.0000

Data Index: 1031017 (C4 (Web Text))

Max Activating Token Index: 1022

Text #13

Max Range: 1.5986. Min Range: -1.5986

Max Act: 0.9658. Min Act: -0.0000

Data Index: 591314 (C4 (Web Text))

Max Activating Token Index: 420

Text #14

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.0220. Min Act: -0.0000

Data Index: 401858 (C4 (Web Text))

Max Activating Token Index: 398

Text #15

Max Range: 1.5986. Min Range: -1.5986

Max Act: 0.9943. Min Act: -0.0000

Data Index: 1080928 (C4 (Web Text))

Max Activating Token Index: 979

Text #16

Max Range: 1.5986. Min Range: -1.5986

Max Act: 0.9862. Min Act: -0.0000

Data Index: 763074 (C4 (Web Text))

Max Activating Token Index: 298

Text #17

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.0227. Min Act: -0.0000

Data Index: 1652730 (Python Code)

Max Activating Token Index: 152

Text #18

Max Range: 1.5986. Min Range: -1.5986

Max Act: 0.9802. Min Act: -0.0000

Data Index: 459757 (C4 (Web Text))

Max Activating Token Index: 884

Text #19

Max Range: 1.5986. Min Range: -1.5986

Max Act: 1.0017. Min Act: -0.0000

Data Index: 1169018 (C4 (Web Text))

Max Activating Token Index: 622