Model: SoLU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 651 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-2l')

Text #0

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0455. Min Act: -0.0002

Data Index: 172679 (C4 (Web Text))

Max Activating Token Index: 520

Text #1

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0425. Min Act: -0.0002

Data Index: 567998 (C4 (Web Text))

Max Activating Token Index: 745

Text #2

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0419. Min Act: -0.0002

Data Index: 402902 (C4 (Web Text))

Max Activating Token Index: 602

Text #3

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0384. Min Act: -0.0002

Data Index: 809036 (C4 (Web Text))

Max Activating Token Index: 201

Text #4

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0368. Min Act: -0.0002

Data Index: 478397 (C4 (Web Text))

Max Activating Token Index: 953

Text #5

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0359. Min Act: -0.0002

Data Index: 1162694 (C4 (Web Text))

Max Activating Token Index: 59

Text #6

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0362. Min Act: -0.0002

Data Index: 302975 (C4 (Web Text))

Max Activating Token Index: 143

Text #7

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0361. Min Act: -0.0002

Data Index: 1296501 (C4 (Web Text))

Max Activating Token Index: 615

Text #8

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0346. Min Act: -0.0002

Data Index: 750974 (C4 (Web Text))

Max Activating Token Index: 964

Text #9

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0344. Min Act: -0.0003

Data Index: 1609010 (Python Code)

Max Activating Token Index: 936

Text #10

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0350. Min Act: -0.0003

Data Index: 361889 (C4 (Web Text))

Max Activating Token Index: 880

Text #11

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0344. Min Act: -0.0003

Data Index: 1633633 (Python Code)

Max Activating Token Index: 65

Text #12

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0338. Min Act: -0.0002

Data Index: 695385 (C4 (Web Text))

Max Activating Token Index: 42

Text #13

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0336. Min Act: -0.0002

Data Index: 600683 (C4 (Web Text))

Max Activating Token Index: 73

Text #14

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0335. Min Act: -0.0003

Data Index: 1579564 (Python Code)

Max Activating Token Index: 31

Text #15

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0333. Min Act: -0.0002

Data Index: 72068 (C4 (Web Text))

Max Activating Token Index: 484

Text #16

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0338. Min Act: -0.0002

Data Index: 300828 (C4 (Web Text))

Max Activating Token Index: 267

Text #17

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0325. Min Act: -0.0003

Data Index: 292251 (C4 (Web Text))

Max Activating Token Index: 840

Text #18

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0330. Min Act: -0.0002

Data Index: 1495943 (Python Code)

Max Activating Token Index: 206

Text #19

Max Range: 0.0455. Min Range: -0.0455

Max Act: 0.0336. Min Act: -0.0002

Data Index: 1351700 (C4 (Web Text))

Max Activating Token Index: 56