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

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

Neuron 372 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: 0.1913. Min Range: -0.1913

Max Act: 0.1913. Min Act: -0.0002

Data Index: 938424 (C4 (Web Text))

Max Activating Token Index: 452

Text #1

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1821. Min Act: -0.0002

Data Index: 8372 (C4 (Web Text))

Max Activating Token Index: 396

Text #2

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1392. Min Act: -0.0002

Data Index: 1319664 (C4 (Web Text))

Max Activating Token Index: 904

Text #3

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1355. Min Act: -0.0002

Data Index: 855172 (C4 (Web Text))

Max Activating Token Index: 985

Text #4

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1265. Min Act: -0.0002

Data Index: 391972 (C4 (Web Text))

Max Activating Token Index: 1007

Text #5

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1313. Min Act: -0.0002

Data Index: 1128700 (C4 (Web Text))

Max Activating Token Index: 713

Text #6

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1275. Min Act: -0.0002

Data Index: 637076 (C4 (Web Text))

Max Activating Token Index: 965

Text #7

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1198. Min Act: -0.0002

Data Index: 595633 (C4 (Web Text))

Max Activating Token Index: 180

Text #8

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1276. Min Act: -0.0002

Data Index: 15762 (C4 (Web Text))

Max Activating Token Index: 864

Text #9

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1199. Min Act: -0.0002

Data Index: 920038 (C4 (Web Text))

Max Activating Token Index: 438

Text #10

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1184. Min Act: -0.0002

Data Index: 1578076 (Python Code)

Max Activating Token Index: 938

Text #11

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1222. Min Act: -0.0002

Data Index: 883756 (C4 (Web Text))

Max Activating Token Index: 781

Text #12

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1171. Min Act: -0.0002

Data Index: 791247 (C4 (Web Text))

Max Activating Token Index: 49

Text #13

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1194. Min Act: -0.0002

Data Index: 1430160 (Python Code)

Max Activating Token Index: 139

Text #14

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1190. Min Act: -0.0002

Data Index: 734819 (C4 (Web Text))

Max Activating Token Index: 942

Text #15

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1145. Min Act: -0.0002

Data Index: 1112221 (C4 (Web Text))

Max Activating Token Index: 890

Text #16

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1063. Min Act: -0.0002

Data Index: 904671 (C4 (Web Text))

Max Activating Token Index: 689

Text #17

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1151. Min Act: -0.0002

Data Index: 1154843 (C4 (Web Text))

Max Activating Token Index: 926

Text #18

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1103. Min Act: -0.0002

Data Index: 1338823 (C4 (Web Text))

Max Activating Token Index: 140

Text #19

Max Range: 0.1913. Min Range: -0.1913

Max Act: 0.1092. Min Act: -0.0002

Data Index: 909875 (C4 (Web Text))

Max Activating Token Index: 382