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

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

Neuron 3174 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: 0.1890. Min Range: -0.1890

Max Act: 0.1890. Min Act: -0.0001

Data Index: 1648091 (Python Code)

Max Activating Token Index: 130

Text #1

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1796. Min Act: -0.0001

Data Index: 1214803 (C4 (Web Text))

Max Activating Token Index: 395

Text #2

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1385. Min Act: -0.0001

Data Index: 616355 (C4 (Web Text))

Max Activating Token Index: 870

Text #3

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1241. Min Act: -0.0001

Data Index: 1128603 (C4 (Web Text))

Max Activating Token Index: 961

Text #4

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1273. Min Act: -0.0001

Data Index: 114800 (C4 (Web Text))

Max Activating Token Index: 625

Text #5

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1169. Min Act: -0.0001

Data Index: 421272 (C4 (Web Text))

Max Activating Token Index: 633

Text #6

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1194. Min Act: -0.0001

Data Index: 663964 (C4 (Web Text))

Max Activating Token Index: 440

Text #7

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1123. Min Act: -0.0001

Data Index: 265474 (C4 (Web Text))

Max Activating Token Index: 856

Text #8

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1200. Min Act: -0.0001

Data Index: 377374 (C4 (Web Text))

Max Activating Token Index: 835

Text #9

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1094. Min Act: -0.0001

Data Index: 695378 (C4 (Web Text))

Max Activating Token Index: 854

Text #10

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1075. Min Act: -0.0001

Data Index: 1289591 (C4 (Web Text))

Max Activating Token Index: 740

Text #11

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1072. Min Act: -0.0001

Data Index: 1601075 (Python Code)

Max Activating Token Index: 10

Text #12

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1179. Min Act: -0.0001

Data Index: 296059 (C4 (Web Text))

Max Activating Token Index: 793

Text #13

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1104. Min Act: -0.0001

Data Index: 1624448 (Python Code)

Max Activating Token Index: 64

Text #14

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1022. Min Act: -0.0001

Data Index: 1570496 (Python Code)

Max Activating Token Index: 1019

Text #15

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1049. Min Act: -0.0001

Data Index: 1242050 (C4 (Web Text))

Max Activating Token Index: 396

Text #16

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1055. Min Act: -0.0001

Data Index: 226948 (C4 (Web Text))

Max Activating Token Index: 128

Text #17

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1017. Min Act: -0.0001

Data Index: 694507 (C4 (Web Text))

Max Activating Token Index: 53

Text #18

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1007. Min Act: -0.0001

Data Index: 1645353 (Python Code)

Max Activating Token Index: 27

Text #19

Max Range: 0.1890. Min Range: -0.1890

Max Act: 0.1034. Min Act: -0.0001

Data Index: 13961 (C4 (Web Text))

Max Activating Token Index: 400