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

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

Neuron 1416 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2856. Min Range: -5.2856

Max Act: 5.2856. Min Act: -0.0001

Data Index: 662331 (C4 (Web Text))

Max Activating Token Index: 1018

Text #1

Max Range: 5.2856. Min Range: -5.2856

Max Act: 5.1017. Min Act: -0.0001

Data Index: 811744 (C4 (Web Text))

Max Activating Token Index: 138

Text #2

Max Range: 5.2856. Min Range: -5.2856

Max Act: 4.1694. Min Act: -0.0001

Data Index: 1323988 (C4 (Web Text))

Max Activating Token Index: 833

Text #3

Max Range: 5.2856. Min Range: -5.2856

Max Act: 4.3973. Min Act: -0.0001

Data Index: 253368 (C4 (Web Text))

Max Activating Token Index: 65

Text #4

Max Range: 5.2856. Min Range: -5.2856

Max Act: 4.1395. Min Act: -0.0001

Data Index: 769107 (C4 (Web Text))

Max Activating Token Index: 193

Text #5

Max Range: 5.2856. Min Range: -5.2856

Max Act: 4.4151. Min Act: -0.0001

Data Index: 664715 (C4 (Web Text))

Max Activating Token Index: 581

Text #6

Max Range: 5.2856. Min Range: -5.2856

Max Act: 4.0495. Min Act: -0.0001

Data Index: 735609 (C4 (Web Text))

Max Activating Token Index: 521

Text #7

Max Range: 5.2856. Min Range: -5.2856

Max Act: 4.0384. Min Act: -0.0001

Data Index: 1025784 (C4 (Web Text))

Max Activating Token Index: 203

Text #8

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.8357. Min Act: -0.0001

Data Index: 973690 (C4 (Web Text))

Max Activating Token Index: 299

Text #9

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.8701. Min Act: -0.0001

Data Index: 1010219 (C4 (Web Text))

Max Activating Token Index: 750

Text #10

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.5089. Min Act: -0.0001

Data Index: 1006475 (C4 (Web Text))

Max Activating Token Index: 721

Text #11

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.4284. Min Act: -0.0001

Data Index: 504253 (C4 (Web Text))

Max Activating Token Index: 809

Text #12

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.3530. Min Act: -0.0001

Data Index: 909648 (C4 (Web Text))

Max Activating Token Index: 161

Text #13

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.3082. Min Act: -0.0001

Data Index: 1342005 (C4 (Web Text))

Max Activating Token Index: 270

Text #14

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.7103. Min Act: -0.0001

Data Index: 201864 (C4 (Web Text))

Max Activating Token Index: 496

Text #15

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.1864. Min Act: -0.0001

Data Index: 693025 (C4 (Web Text))

Max Activating Token Index: 907

Text #16

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.1838. Min Act: -0.0001

Data Index: 1206593 (C4 (Web Text))

Max Activating Token Index: 872

Text #17

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.3166. Min Act: -0.0001

Data Index: 508269 (C4 (Web Text))

Max Activating Token Index: 240

Text #18

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.3211. Min Act: -0.0001

Data Index: 387562 (C4 (Web Text))

Max Activating Token Index: 163

Text #19

Max Range: 5.2856. Min Range: -5.2856

Max Act: 3.0011. Min Act: -0.0001

Data Index: 1162905 (C4 (Web Text))

Max Activating Token Index: 311