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

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

Neuron 3622 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.5144. Min Act: -0.0000

Data Index: 814111 (C4 (Web Text))

Max Activating Token Index: 628

Text #1

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.5123. Min Act: -0.0000

Data Index: 676967 (C4 (Web Text))

Max Activating Token Index: 726

Text #2

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4667. Min Act: -0.0000

Data Index: 1148665 (C4 (Web Text))

Max Activating Token Index: 835

Text #3

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4795. Min Act: -0.0000

Data Index: 763233 (C4 (Web Text))

Max Activating Token Index: 845

Text #4

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4498. Min Act: -0.0000

Data Index: 13709 (C4 (Web Text))

Max Activating Token Index: 385

Text #5

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4385. Min Act: -0.0000

Data Index: 823514 (C4 (Web Text))

Max Activating Token Index: 509

Text #6

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4449. Min Act: -0.0000

Data Index: 159148 (C4 (Web Text))

Max Activating Token Index: 579

Text #7

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4262. Min Act: -0.0000

Data Index: 1105288 (C4 (Web Text))

Max Activating Token Index: 140

Text #8

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.3931. Min Act: -0.0000

Data Index: 20113 (C4 (Web Text))

Max Activating Token Index: 558

Text #9

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4156. Min Act: -0.0000

Data Index: 498877 (C4 (Web Text))

Max Activating Token Index: 666

Text #10

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4133. Min Act: -0.0000

Data Index: 870844 (C4 (Web Text))

Max Activating Token Index: 859

Text #11

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4105. Min Act: -0.0000

Data Index: 907780 (C4 (Web Text))

Max Activating Token Index: 191

Text #12

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4054. Min Act: -0.0000

Data Index: 267212 (C4 (Web Text))

Max Activating Token Index: 696

Text #13

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4050. Min Act: -0.0000

Data Index: 690638 (C4 (Web Text))

Max Activating Token Index: 618

Text #14

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.3662. Min Act: -0.0000

Data Index: 128015 (C4 (Web Text))

Max Activating Token Index: 707

Text #15

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.3911. Min Act: -0.0000

Data Index: 188002 (C4 (Web Text))

Max Activating Token Index: 435

Text #16

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.3739. Min Act: -0.0000

Data Index: 1367422 (Python Code)

Max Activating Token Index: 521

Text #17

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.4240. Min Act: -0.0000

Data Index: 1020115 (C4 (Web Text))

Max Activating Token Index: 884

Text #18

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.3437. Min Act: -0.0000

Data Index: 367066 (C4 (Web Text))

Max Activating Token Index: 353

Text #19

Max Range: 0.5144. Min Range: -0.5144

Max Act: 0.3850. Min Act: -0.0000

Data Index: 552004 (C4 (Web Text))

Max Activating Token Index: 185