Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4928 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.6476. Min Act: -0.0000

Data Index: 1383289 (Python Code)

Max Activating Token Index: 39

Text #1

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.6640. Min Act: -0.0000

Data Index: 1502482 (Python Code)

Max Activating Token Index: 30

Text #2

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.5792. Min Act: -0.0000

Data Index: 708207 (C4 (Web Text))

Max Activating Token Index: 856

Text #3

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.5740. Min Act: -0.0000

Data Index: 1606693 (Python Code)

Max Activating Token Index: 28

Text #4

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.5398. Min Act: -0.0000

Data Index: 1467004 (Python Code)

Max Activating Token Index: 47

Text #5

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4487. Min Act: -0.0000

Data Index: 857256 (C4 (Web Text))

Max Activating Token Index: 229

Text #6

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4521. Min Act: -0.0000

Data Index: 1087745 (C4 (Web Text))

Max Activating Token Index: 300

Text #7

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4961. Min Act: -0.0000

Data Index: 352339 (C4 (Web Text))

Max Activating Token Index: 329

Text #8

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4730. Min Act: -0.0000

Data Index: 1376827 (Python Code)

Max Activating Token Index: 142

Text #9

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4457. Min Act: -0.0000

Data Index: 1557905 (Python Code)

Max Activating Token Index: 44

Text #10

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4443. Min Act: -0.0000

Data Index: 707152 (C4 (Web Text))

Max Activating Token Index: 88

Text #11

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4329. Min Act: -0.0000

Data Index: 56289 (C4 (Web Text))

Max Activating Token Index: 592

Text #12

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.3994. Min Act: -0.0000

Data Index: 677288 (C4 (Web Text))

Max Activating Token Index: 449

Text #13

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.3852. Min Act: -0.0000

Data Index: 525407 (C4 (Web Text))

Max Activating Token Index: 768

Text #14

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4048. Min Act: -0.0000

Data Index: 1414181 (Python Code)

Max Activating Token Index: 953

Text #15

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4047. Min Act: -0.0000

Data Index: 357131 (C4 (Web Text))

Max Activating Token Index: 160

Text #16

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4036. Min Act: -0.0000

Data Index: 1460756 (Python Code)

Max Activating Token Index: 523

Text #17

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.3872. Min Act: -0.0000

Data Index: 50899 (C4 (Web Text))

Max Activating Token Index: 342

Text #18

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.4101. Min Act: -0.0000

Data Index: 1520548 (Python Code)

Max Activating Token Index: 36

Text #19

Max Range: 0.6640. Min Range: -0.6640

Max Act: 0.3907. Min Act: -0.0000

Data Index: 1581386 (Python Code)

Max Activating Token Index: 562