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

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

Neuron 1654 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2317. Min Range: -3.2317

Max Act: 3.2317. Min Act: -0.0000

Data Index: 854998 (C4 (Web Text))

Max Activating Token Index: 414

Text #1

Max Range: 3.2317. Min Range: -3.2317

Max Act: 3.0801. Min Act: -0.0000

Data Index: 1610165 (Python Code)

Max Activating Token Index: 56

Text #2

Max Range: 3.2317. Min Range: -3.2317

Max Act: 3.1414. Min Act: -0.0000

Data Index: 923307 (C4 (Web Text))

Max Activating Token Index: 123

Text #3

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.7296. Min Act: -0.0000

Data Index: 80091 (C4 (Web Text))

Max Activating Token Index: 524

Text #4

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.6806. Min Act: -0.0000

Data Index: 227034 (C4 (Web Text))

Max Activating Token Index: 146

Text #5

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.5932. Min Act: -0.0000

Data Index: 1411343 (Python Code)

Max Activating Token Index: 1018

Text #6

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.5517. Min Act: -0.0000

Data Index: 1298374 (C4 (Web Text))

Max Activating Token Index: 726

Text #7

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.3219. Min Act: -0.0000

Data Index: 826701 (C4 (Web Text))

Max Activating Token Index: 367

Text #8

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.7640. Min Act: -0.0000

Data Index: 231235 (C4 (Web Text))

Max Activating Token Index: 743

Text #9

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.3612. Min Act: -0.0000

Data Index: 136580 (C4 (Web Text))

Max Activating Token Index: 550

Text #10

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.3554. Min Act: -0.0000

Data Index: 1398424 (Python Code)

Max Activating Token Index: 183

Text #11

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.2147. Min Act: -0.0000

Data Index: 1550972 (Python Code)

Max Activating Token Index: 69

Text #12

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.3319. Min Act: -0.0000

Data Index: 1544268 (Python Code)

Max Activating Token Index: 713

Text #13

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.2916. Min Act: -0.0000

Data Index: 60541 (C4 (Web Text))

Max Activating Token Index: 240

Text #14

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.4925. Min Act: -0.0000

Data Index: 672149 (C4 (Web Text))

Max Activating Token Index: 670

Text #15

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.2405. Min Act: -0.0000

Data Index: 1523421 (Python Code)

Max Activating Token Index: 27

Text #16

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.0126. Min Act: -0.0000

Data Index: 1100956 (C4 (Web Text))

Max Activating Token Index: 655

Text #17

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.0900. Min Act: -0.0000

Data Index: 1505169 (Python Code)

Max Activating Token Index: 907

Text #18

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.4061. Min Act: -0.0000

Data Index: 385087 (C4 (Web Text))

Max Activating Token Index: 405

Text #19

Max Range: 3.2317. Min Range: -3.2317

Max Act: 2.0685. Min Act: -0.0000

Data Index: 1361894 (Python Code)

Max Activating Token Index: 38