Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 370 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.6138. Min Act: -0.0001

Data Index: 287327 (C4 (Web Text))

Max Activating Token Index: 909

Text #1

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.6101. Min Act: -0.0001

Data Index: 749928 (C4 (Web Text))

Max Activating Token Index: 113

Text #2

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.6094. Min Act: -0.0001

Data Index: 1364625 (Python Code)

Max Activating Token Index: 944

Text #3

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.6043. Min Act: -0.0001

Data Index: 883349 (C4 (Web Text))

Max Activating Token Index: 1002

Text #4

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.6209. Min Act: -0.0001

Data Index: 255061 (C4 (Web Text))

Max Activating Token Index: 1013

Text #5

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.6138. Min Act: -0.0001

Data Index: 1635802 (Python Code)

Max Activating Token Index: 32

Text #6

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5858. Min Act: -0.0001

Data Index: 377712 (C4 (Web Text))

Max Activating Token Index: 518

Text #7

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.6126. Min Act: -0.0001

Data Index: 1320460 (C4 (Web Text))

Max Activating Token Index: 314

Text #8

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.6010. Min Act: -0.0001

Data Index: 221861 (C4 (Web Text))

Max Activating Token Index: 205

Text #9

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5847. Min Act: -0.0001

Data Index: 528985 (C4 (Web Text))

Max Activating Token Index: 43

Text #10

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.6208. Min Act: -0.0001

Data Index: 370318 (C4 (Web Text))

Max Activating Token Index: 195

Text #11

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5795. Min Act: -0.0001

Data Index: 1379251 (Python Code)

Max Activating Token Index: 1016

Text #12

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5422. Min Act: -0.0001

Data Index: 1295505 (C4 (Web Text))

Max Activating Token Index: 924

Text #13

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5730. Min Act: -0.0001

Data Index: 1068811 (C4 (Web Text))

Max Activating Token Index: 420

Text #14

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5605. Min Act: -0.0001

Data Index: 938119 (C4 (Web Text))

Max Activating Token Index: 837

Text #15

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5729. Min Act: -0.0001

Data Index: 391809 (C4 (Web Text))

Max Activating Token Index: 781

Text #16

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5733. Min Act: -0.0001

Data Index: 880927 (C4 (Web Text))

Max Activating Token Index: 35

Text #17

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5565. Min Act: -0.0001

Data Index: 834725 (C4 (Web Text))

Max Activating Token Index: 647

Text #18

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5846. Min Act: -0.0001

Data Index: 534780 (C4 (Web Text))

Max Activating Token Index: 497

Text #19

Max Range: 0.6209. Min Range: -0.6209

Max Act: 0.5454. Min Act: -0.0001

Data Index: 1334417 (C4 (Web Text))

Max Activating Token Index: 889