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

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

Neuron 1446 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2575. Min Act: -0.0005

Data Index: 23350 (C4 (Web Text))

Max Activating Token Index: 950

Text #1

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2645. Min Act: -0.0005

Data Index: 186190 (C4 (Web Text))

Max Activating Token Index: 302

Text #2

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2262. Min Act: -0.0005

Data Index: 3107 (C4 (Web Text))

Max Activating Token Index: 700

Text #3

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2395. Min Act: -0.0005

Data Index: 345173 (C4 (Web Text))

Max Activating Token Index: 987

Text #4

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2429. Min Act: -0.0005

Data Index: 1009191 (C4 (Web Text))

Max Activating Token Index: 855

Text #5

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2240. Min Act: -0.0005

Data Index: 1327659 (C4 (Web Text))

Max Activating Token Index: 845

Text #6

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2307. Min Act: -0.0005

Data Index: 754782 (C4 (Web Text))

Max Activating Token Index: 515

Text #7

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2199. Min Act: -0.0005

Data Index: 151901 (C4 (Web Text))

Max Activating Token Index: 8

Text #8

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2276. Min Act: -0.0005

Data Index: 922918 (C4 (Web Text))

Max Activating Token Index: 1012

Text #9

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2217. Min Act: -0.0005

Data Index: 1112438 (C4 (Web Text))

Max Activating Token Index: 288

Text #10

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2383. Min Act: -0.0005

Data Index: 6470 (C4 (Web Text))

Max Activating Token Index: 778

Text #11

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2137. Min Act: -0.0005

Data Index: 151579 (C4 (Web Text))

Max Activating Token Index: 515

Text #12

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2259. Min Act: -0.0005

Data Index: 1004169 (C4 (Web Text))

Max Activating Token Index: 763

Text #13

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2092. Min Act: -0.0005

Data Index: 555260 (C4 (Web Text))

Max Activating Token Index: 112

Text #14

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2136. Min Act: -0.0005

Data Index: 810659 (C4 (Web Text))

Max Activating Token Index: 271

Text #15

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2079. Min Act: -0.0005

Data Index: 919885 (C4 (Web Text))

Max Activating Token Index: 951

Text #16

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2167. Min Act: -0.0005

Data Index: 675521 (C4 (Web Text))

Max Activating Token Index: 136

Text #17

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.1948. Min Act: -0.0005

Data Index: 369087 (C4 (Web Text))

Max Activating Token Index: 484

Text #18

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.2187. Min Act: -0.0005

Data Index: 1176552 (C4 (Web Text))

Max Activating Token Index: 887

Text #19

Max Range: 0.2645. Min Range: -0.2645

Max Act: 0.1992. Min Act: -0.0005

Data Index: 811974 (C4 (Web Text))

Max Activating Token Index: 1023