Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2020 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1591. Min Act: -0.0001

Data Index: 1276008 (C4 (Web Text))

Max Activating Token Index: 563

Text #1

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1474. Min Act: -0.0001

Data Index: 469433 (C4 (Web Text))

Max Activating Token Index: 460

Text #2

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1330. Min Act: -0.0001

Data Index: 654552 (C4 (Web Text))

Max Activating Token Index: 606

Text #3

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1365. Min Act: -0.0001

Data Index: 49198 (C4 (Web Text))

Max Activating Token Index: 994

Text #4

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1294. Min Act: -0.0000

Data Index: 625479 (C4 (Web Text))

Max Activating Token Index: 290

Text #5

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1291. Min Act: -0.0001

Data Index: 461735 (C4 (Web Text))

Max Activating Token Index: 157

Text #6

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1222. Min Act: -0.0001

Data Index: 1295874 (C4 (Web Text))

Max Activating Token Index: 56

Text #7

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1309. Min Act: -0.0001

Data Index: 1106501 (C4 (Web Text))

Max Activating Token Index: 625

Text #8

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1260. Min Act: -0.0001

Data Index: 19034 (C4 (Web Text))

Max Activating Token Index: 455

Text #9

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1257. Min Act: -0.0001

Data Index: 1341123 (C4 (Web Text))

Max Activating Token Index: 90

Text #10

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1242. Min Act: -0.0001

Data Index: 119174 (C4 (Web Text))

Max Activating Token Index: 477

Text #11

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1209. Min Act: -0.0001

Data Index: 546529 (C4 (Web Text))

Max Activating Token Index: 885

Text #12

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1148. Min Act: -0.0001

Data Index: 65482 (C4 (Web Text))

Max Activating Token Index: 935

Text #13

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1165. Min Act: -0.0001

Data Index: 603940 (C4 (Web Text))

Max Activating Token Index: 598

Text #14

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1129. Min Act: -0.0001

Data Index: 344639 (C4 (Web Text))

Max Activating Token Index: 419

Text #15

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1115. Min Act: -0.0001

Data Index: 1152088 (C4 (Web Text))

Max Activating Token Index: 391

Text #16

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1054. Min Act: -0.0001

Data Index: 750978 (C4 (Web Text))

Max Activating Token Index: 416

Text #17

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1094. Min Act: -0.0001

Data Index: 324215 (C4 (Web Text))

Max Activating Token Index: 67

Text #18

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1074. Min Act: -0.0001

Data Index: 1175379 (C4 (Web Text))

Max Activating Token Index: 531

Text #19

Max Range: 0.1591. Min Range: -0.1591

Max Act: 0.1043. Min Act: -0.0001

Data Index: 1386250 (Python Code)

Max Activating Token Index: 173