Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5277 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1631. Min Act: -0.0001

Data Index: 91105 (C4 (Web Text))

Max Activating Token Index: 265

Text #1

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1436. Min Act: -0.0001

Data Index: 412037 (C4 (Web Text))

Max Activating Token Index: 397

Text #2

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1336. Min Act: -0.0001

Data Index: 796299 (C4 (Web Text))

Max Activating Token Index: 820

Text #3

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1277. Min Act: -0.0001

Data Index: 349473 (C4 (Web Text))

Max Activating Token Index: 901

Text #4

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1145. Min Act: -0.0001

Data Index: 375582 (C4 (Web Text))

Max Activating Token Index: 676

Text #5

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1128. Min Act: -0.0001

Data Index: 1199641 (C4 (Web Text))

Max Activating Token Index: 202

Text #6

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1195. Min Act: -0.0001

Data Index: 1260275 (C4 (Web Text))

Max Activating Token Index: 331

Text #7

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1130. Min Act: -0.0001

Data Index: 131028 (C4 (Web Text))

Max Activating Token Index: 230

Text #8

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1054. Min Act: -0.0001

Data Index: 15177 (C4 (Web Text))

Max Activating Token Index: 858

Text #9

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.1043. Min Act: -0.0001

Data Index: 1351586 (C4 (Web Text))

Max Activating Token Index: 896

Text #10

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0949. Min Act: -0.0001

Data Index: 1045890 (C4 (Web Text))

Max Activating Token Index: 823

Text #11

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0949. Min Act: -0.0001

Data Index: 401611 (C4 (Web Text))

Max Activating Token Index: 254

Text #12

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0972. Min Act: -0.0001

Data Index: 1195941 (C4 (Web Text))

Max Activating Token Index: 221

Text #13

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0926. Min Act: -0.0001

Data Index: 1506168 (Python Code)

Max Activating Token Index: 645

Text #14

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0966. Min Act: -0.0001

Data Index: 462055 (C4 (Web Text))

Max Activating Token Index: 390

Text #15

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0923. Min Act: -0.0001

Data Index: 463914 (C4 (Web Text))

Max Activating Token Index: 453

Text #16

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0956. Min Act: -0.0001

Data Index: 825608 (C4 (Web Text))

Max Activating Token Index: 498

Text #17

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0950. Min Act: -0.0001

Data Index: 734030 (C4 (Web Text))

Max Activating Token Index: 175

Text #18

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0956. Min Act: -0.0001

Data Index: 238921 (C4 (Web Text))

Max Activating Token Index: 352

Text #19

Max Range: 0.1631. Min Range: -0.1631

Max Act: 0.0983. Min Act: -0.0001

Data Index: 640058 (C4 (Web Text))

Max Activating Token Index: 57