Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 615 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.4830. Min Act: -0.0006

Data Index: 113369 (C4 (Web Text))

Max Activating Token Index: 255

Text #1

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5608. Min Act: -0.0006

Data Index: 1140486 (C4 (Web Text))

Max Activating Token Index: 401

Text #2

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5751. Min Act: -0.0006

Data Index: 816314 (C4 (Web Text))

Max Activating Token Index: 1017

Text #3

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5549. Min Act: -0.0006

Data Index: 869705 (C4 (Web Text))

Max Activating Token Index: 991

Text #4

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5932. Min Act: -0.0006

Data Index: 1069843 (C4 (Web Text))

Max Activating Token Index: 142

Text #5

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5748. Min Act: -0.0006

Data Index: 510550 (C4 (Web Text))

Max Activating Token Index: 807

Text #6

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5395. Min Act: -0.0006

Data Index: 1288075 (C4 (Web Text))

Max Activating Token Index: 565

Text #7

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5555. Min Act: -0.0006

Data Index: 1121564 (C4 (Web Text))

Max Activating Token Index: 81

Text #8

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5331. Min Act: -0.0006

Data Index: 464932 (C4 (Web Text))

Max Activating Token Index: 854

Text #9

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5337. Min Act: -0.0006

Data Index: 310838 (C4 (Web Text))

Max Activating Token Index: 818

Text #10

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5497. Min Act: -0.0006

Data Index: 769240 (C4 (Web Text))

Max Activating Token Index: 4

Text #11

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5484. Min Act: -0.0006

Data Index: 1028839 (C4 (Web Text))

Max Activating Token Index: 276

Text #12

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5452. Min Act: -0.0006

Data Index: 47201 (C4 (Web Text))

Max Activating Token Index: 602

Text #13

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5765. Min Act: -0.0006

Data Index: 488699 (C4 (Web Text))

Max Activating Token Index: 969

Text #14

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5407. Min Act: -0.0006

Data Index: 1196260 (C4 (Web Text))

Max Activating Token Index: 224

Text #15

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5357. Min Act: -0.0006

Data Index: 1174287 (C4 (Web Text))

Max Activating Token Index: 37

Text #16

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5353. Min Act: -0.0006

Data Index: 10916 (C4 (Web Text))

Max Activating Token Index: 222

Text #17

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5164. Min Act: -0.0006

Data Index: 1265629 (C4 (Web Text))

Max Activating Token Index: 766

Text #18

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.5150. Min Act: -0.0006

Data Index: 30659 (C4 (Web Text))

Max Activating Token Index: 526

Text #19

Max Range: 0.5932. Min Range: -0.5932

Max Act: 0.4987. Min Act: -0.0006

Data Index: 2295 (C4 (Web Text))

Max Activating Token Index: 968