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

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

Neuron 3839 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.4397. Min Act: -0.0001

Data Index: 442877 (C4 (Web Text))

Max Activating Token Index: 298

Text #1

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.3674. Min Act: -0.0001

Data Index: 259596 (C4 (Web Text))

Max Activating Token Index: 530

Text #2

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.2560. Min Act: -0.0001

Data Index: 99889 (C4 (Web Text))

Max Activating Token Index: 173

Text #3

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.2045. Min Act: -0.0001

Data Index: 1534332 (Python Code)

Max Activating Token Index: 850

Text #4

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.2034. Min Act: -0.0001

Data Index: 1480006 (Python Code)

Max Activating Token Index: 841

Text #5

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.1526. Min Act: -0.0001

Data Index: 501801 (C4 (Web Text))

Max Activating Token Index: 613

Text #6

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.0929. Min Act: -0.0000

Data Index: 1505876 (Python Code)

Max Activating Token Index: 852

Text #7

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.1388. Min Act: -0.0000

Data Index: 1441178 (Python Code)

Max Activating Token Index: 751

Text #8

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.1755. Min Act: -0.0001

Data Index: 1221309 (C4 (Web Text))

Max Activating Token Index: 675

Text #9

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.2477. Min Act: -0.0001

Data Index: 675588 (C4 (Web Text))

Max Activating Token Index: 179

Text #10

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.1984. Min Act: -0.0000

Data Index: 343524 (C4 (Web Text))

Max Activating Token Index: 458

Text #11

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.1103. Min Act: -0.0001

Data Index: 339702 (C4 (Web Text))

Max Activating Token Index: 117

Text #12

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.1872. Min Act: -0.0000

Data Index: 1408451 (Python Code)

Max Activating Token Index: 1008

Text #13

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.1749. Min Act: -0.0001

Data Index: 29502 (C4 (Web Text))

Max Activating Token Index: 1021

Text #14

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.1692. Min Act: -0.0001

Data Index: 1457805 (Python Code)

Max Activating Token Index: 865

Text #15

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.0978. Min Act: -0.0001

Data Index: 1480987 (Python Code)

Max Activating Token Index: 158

Text #16

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.0153. Min Act: -0.0000

Data Index: 1108676 (C4 (Web Text))

Max Activating Token Index: 886

Text #17

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.1863. Min Act: -0.0001

Data Index: 515495 (C4 (Web Text))

Max Activating Token Index: 405

Text #18

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.0909. Min Act: -0.0001

Data Index: 1321196 (C4 (Web Text))

Max Activating Token Index: 415

Text #19

Max Range: 1.4397. Min Range: -1.4397

Max Act: 1.0557. Min Act: -0.0001

Data Index: 97857 (C4 (Web Text))

Max Activating Token Index: 445