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

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

Neuron 427 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: 0.4479. Min Range: -0.4479

Max Act: 0.4479. Min Act: -0.0002

Data Index: 47878 (C4 (Web Text))

Max Activating Token Index: 550

Text #1

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.4240. Min Act: -0.0002

Data Index: 451272 (C4 (Web Text))

Max Activating Token Index: 327

Text #2

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.4235. Min Act: -0.0002

Data Index: 1063163 (C4 (Web Text))

Max Activating Token Index: 280

Text #3

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.3231. Min Act: -0.0002

Data Index: 314822 (C4 (Web Text))

Max Activating Token Index: 823

Text #4

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.3284. Min Act: -0.0002

Data Index: 1383581 (Python Code)

Max Activating Token Index: 928

Text #5

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.3435. Min Act: -0.0002

Data Index: 1105038 (C4 (Web Text))

Max Activating Token Index: 880

Text #6

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.3080. Min Act: -0.0002

Data Index: 1411172 (Python Code)

Max Activating Token Index: 227

Text #7

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.2563. Min Act: -0.0002

Data Index: 780700 (C4 (Web Text))

Max Activating Token Index: 247

Text #8

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.2387. Min Act: -0.0001

Data Index: 1281704 (C4 (Web Text))

Max Activating Token Index: 335

Text #9

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.2680. Min Act: -0.0002

Data Index: 1279514 (C4 (Web Text))

Max Activating Token Index: 319

Text #10

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.2449. Min Act: -0.0002

Data Index: 1572470 (Python Code)

Max Activating Token Index: 607

Text #11

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.2933. Min Act: -0.0002

Data Index: 921074 (C4 (Web Text))

Max Activating Token Index: 91

Text #12

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.2100. Min Act: -0.0002

Data Index: 1422031 (Python Code)

Max Activating Token Index: 232

Text #13

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.2206. Min Act: -0.0002

Data Index: 422869 (C4 (Web Text))

Max Activating Token Index: 99

Text #14

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.1790. Min Act: -0.0002

Data Index: 1287821 (C4 (Web Text))

Max Activating Token Index: 614

Text #15

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.1820. Min Act: -0.0001

Data Index: 1099199 (C4 (Web Text))

Max Activating Token Index: 260

Text #16

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.1937. Min Act: -0.0002

Data Index: 1400932 (Python Code)

Max Activating Token Index: 326

Text #17

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.1804. Min Act: -0.0002

Data Index: 1453385 (Python Code)

Max Activating Token Index: 237

Text #18

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.1451. Min Act: -0.0002

Data Index: 1427905 (Python Code)

Max Activating Token Index: 185

Text #19

Max Range: 0.4479. Min Range: -0.4479

Max Act: 0.1531. Min Act: -0.0002

Data Index: 433632 (C4 (Web Text))

Max Activating Token Index: 599