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

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

Neuron 5577 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.8796. Min Act: -0.0000

Data Index: 420388 (C4 (Web Text))

Max Activating Token Index: 870

Text #1

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.8259. Min Act: -0.0000

Data Index: 187654 (C4 (Web Text))

Max Activating Token Index: 811

Text #2

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.6502. Min Act: -0.0000

Data Index: 284573 (C4 (Web Text))

Max Activating Token Index: 993

Text #3

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.8001. Min Act: -0.0000

Data Index: 47771 (C4 (Web Text))

Max Activating Token Index: 678

Text #4

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.4256. Min Act: -0.0000

Data Index: 1526237 (Python Code)

Max Activating Token Index: 394

Text #5

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.4406. Min Act: -0.0000

Data Index: 577625 (C4 (Web Text))

Max Activating Token Index: 300

Text #6

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.1350. Min Act: -0.0000

Data Index: 1649788 (Python Code)

Max Activating Token Index: 251

Text #7

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.3540. Min Act: -0.0000

Data Index: 188112 (C4 (Web Text))

Max Activating Token Index: 649

Text #8

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.3571. Min Act: -0.0000

Data Index: 1542931 (Python Code)

Max Activating Token Index: 279

Text #9

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.5819. Min Act: -0.0000

Data Index: 478325 (C4 (Web Text))

Max Activating Token Index: 772

Text #10

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.3470. Min Act: -0.0000

Data Index: 1481093 (Python Code)

Max Activating Token Index: 325

Text #11

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.4394. Min Act: -0.0000

Data Index: 1291622 (C4 (Web Text))

Max Activating Token Index: 889

Text #12

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.2862. Min Act: -0.0000

Data Index: 507147 (C4 (Web Text))

Max Activating Token Index: 790

Text #13

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.5177. Min Act: -0.0000

Data Index: 891789 (C4 (Web Text))

Max Activating Token Index: 505

Text #14

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.2763. Min Act: -0.0000

Data Index: 64121 (C4 (Web Text))

Max Activating Token Index: 475

Text #15

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.2480. Min Act: -0.0000

Data Index: 523190 (C4 (Web Text))

Max Activating Token Index: 636

Text #16

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.3366. Min Act: -0.0000

Data Index: 1083886 (C4 (Web Text))

Max Activating Token Index: 494

Text #17

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.2025. Min Act: -0.0000

Data Index: 486260 (C4 (Web Text))

Max Activating Token Index: 650

Text #18

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.0850. Min Act: -0.0000

Data Index: 1341800 (C4 (Web Text))

Max Activating Token Index: 570

Text #19

Max Range: 2.8796. Min Range: -2.8796

Max Act: 2.2946. Min Act: -0.0000

Data Index: 184230 (C4 (Web Text))

Max Activating Token Index: 221