Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 680 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0692. Min Act: -0.0006

Data Index: 1682 (C4 (Web Text))

Max Activating Token Index: 17

Text #1

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0768. Min Act: -0.0005

Data Index: 335065 (C4 (Web Text))

Max Activating Token Index: 288

Text #2

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0663. Min Act: -0.0006

Data Index: 1127406 (C4 (Web Text))

Max Activating Token Index: 268

Text #3

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0651. Min Act: -0.0005

Data Index: 912095 (C4 (Web Text))

Max Activating Token Index: 114

Text #4

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0618. Min Act: -0.0006

Data Index: 1114941 (C4 (Web Text))

Max Activating Token Index: 266

Text #5

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0614. Min Act: -0.0006

Data Index: 1144263 (C4 (Web Text))

Max Activating Token Index: 922

Text #6

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0596. Min Act: -0.0005

Data Index: 1210348 (C4 (Web Text))

Max Activating Token Index: 350

Text #7

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0618. Min Act: -0.0006

Data Index: 946011 (C4 (Web Text))

Max Activating Token Index: 359

Text #8

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0610. Min Act: -0.0005

Data Index: 1044590 (C4 (Web Text))

Max Activating Token Index: 286

Text #9

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0586. Min Act: -0.0006

Data Index: 599706 (C4 (Web Text))

Max Activating Token Index: 28

Text #10

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0582. Min Act: -0.0006

Data Index: 763569 (C4 (Web Text))

Max Activating Token Index: 612

Text #11

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0629. Min Act: -0.0006

Data Index: 473171 (C4 (Web Text))

Max Activating Token Index: 754

Text #12

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0596. Min Act: -0.0006

Data Index: 776262 (C4 (Web Text))

Max Activating Token Index: 871

Text #13

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0586. Min Act: -0.0006

Data Index: 138371 (C4 (Web Text))

Max Activating Token Index: 90

Text #14

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0585. Min Act: -0.0006

Data Index: 457159 (C4 (Web Text))

Max Activating Token Index: 11

Text #15

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0589. Min Act: -0.0006

Data Index: 894418 (C4 (Web Text))

Max Activating Token Index: 684

Text #16

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0555. Min Act: -0.0006

Data Index: 145638 (C4 (Web Text))

Max Activating Token Index: 951

Text #17

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0553. Min Act: -0.0006

Data Index: 945161 (C4 (Web Text))

Max Activating Token Index: 812

Text #18

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0562. Min Act: -0.0005

Data Index: 324295 (C4 (Web Text))

Max Activating Token Index: 452

Text #19

Max Range: 0.0768. Min Range: -0.0768

Max Act: 0.0545. Min Act: -0.0005

Data Index: 548178 (C4 (Web Text))

Max Activating Token Index: 213