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

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

Neuron 2038 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.5454. Min Act: -0.0001

Data Index: 613255 (C4 (Web Text))

Max Activating Token Index: 726

Text #1

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.5188. Min Act: -0.0001

Data Index: 902098 (C4 (Web Text))

Max Activating Token Index: 256

Text #2

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4255. Min Act: -0.0001

Data Index: 713395 (C4 (Web Text))

Max Activating Token Index: 720

Text #3

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4383. Min Act: -0.0001

Data Index: 1111203 (C4 (Web Text))

Max Activating Token Index: 931

Text #4

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4079. Min Act: -0.0001

Data Index: 450056 (C4 (Web Text))

Max Activating Token Index: 667

Text #5

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4526. Min Act: -0.0002

Data Index: 338322 (C4 (Web Text))

Max Activating Token Index: 1010

Text #6

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4208. Min Act: -0.0001

Data Index: 1045978 (C4 (Web Text))

Max Activating Token Index: 116

Text #7

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.3947. Min Act: -0.0001

Data Index: 972282 (C4 (Web Text))

Max Activating Token Index: 798

Text #8

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4522. Min Act: -0.0001

Data Index: 462233 (C4 (Web Text))

Max Activating Token Index: 976

Text #9

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.3973. Min Act: -0.0001

Data Index: 1229775 (C4 (Web Text))

Max Activating Token Index: 681

Text #10

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4088. Min Act: -0.0002

Data Index: 760100 (C4 (Web Text))

Max Activating Token Index: 802

Text #11

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4175. Min Act: -0.0001

Data Index: 200052 (C4 (Web Text))

Max Activating Token Index: 940

Text #12

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4041. Min Act: -0.0001

Data Index: 489807 (C4 (Web Text))

Max Activating Token Index: 998

Text #13

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4230. Min Act: -0.0002

Data Index: 529294 (C4 (Web Text))

Max Activating Token Index: 330

Text #14

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4246. Min Act: -0.0001

Data Index: 233980 (C4 (Web Text))

Max Activating Token Index: 168

Text #15

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4122. Min Act: -0.0001

Data Index: 990017 (C4 (Web Text))

Max Activating Token Index: 517

Text #16

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.3843. Min Act: -0.0002

Data Index: 819219 (C4 (Web Text))

Max Activating Token Index: 463

Text #17

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.3966. Min Act: -0.0002

Data Index: 491491 (C4 (Web Text))

Max Activating Token Index: 786

Text #18

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.4156. Min Act: -0.0001

Data Index: 1027839 (C4 (Web Text))

Max Activating Token Index: 416

Text #19

Max Range: 0.5454. Min Range: -0.5454

Max Act: 0.3831. Min Act: -0.0001

Data Index: 960148 (C4 (Web Text))

Max Activating Token Index: 957