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

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

Neuron 1646 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: 0.1730. Min Range: -0.1730

Max Act: 0.1730. Min Act: -0.0001

Data Index: 527378 (C4 (Web Text))

Max Activating Token Index: 306

Text #1

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1567. Min Act: -0.0002

Data Index: 926288 (C4 (Web Text))

Max Activating Token Index: 368

Text #2

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1605. Min Act: -0.0002

Data Index: 227977 (C4 (Web Text))

Max Activating Token Index: 166

Text #3

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1585. Min Act: -0.0002

Data Index: 190107 (C4 (Web Text))

Max Activating Token Index: 634

Text #4

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1542. Min Act: -0.0001

Data Index: 1051264 (C4 (Web Text))

Max Activating Token Index: 181

Text #5

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1545. Min Act: -0.0002

Data Index: 740072 (C4 (Web Text))

Max Activating Token Index: 955

Text #6

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1520. Min Act: -0.0002

Data Index: 1293355 (C4 (Web Text))

Max Activating Token Index: 745

Text #7

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1527. Min Act: -0.0002

Data Index: 209200 (C4 (Web Text))

Max Activating Token Index: 288

Text #8

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1485. Min Act: -0.0001

Data Index: 819878 (C4 (Web Text))

Max Activating Token Index: 168

Text #9

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1499. Min Act: -0.0002

Data Index: 1184704 (C4 (Web Text))

Max Activating Token Index: 250

Text #10

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1555. Min Act: -0.0002

Data Index: 901582 (C4 (Web Text))

Max Activating Token Index: 458

Text #11

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1503. Min Act: -0.0002

Data Index: 206366 (C4 (Web Text))

Max Activating Token Index: 220

Text #12

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1500. Min Act: -0.0002

Data Index: 1015937 (C4 (Web Text))

Max Activating Token Index: 541

Text #13

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1444. Min Act: -0.0002

Data Index: 181622 (C4 (Web Text))

Max Activating Token Index: 923

Text #14

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1526. Min Act: -0.0002

Data Index: 999164 (C4 (Web Text))

Max Activating Token Index: 427

Text #15

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1452. Min Act: -0.0001

Data Index: 195293 (C4 (Web Text))

Max Activating Token Index: 81

Text #16

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1431. Min Act: -0.0001

Data Index: 358795 (C4 (Web Text))

Max Activating Token Index: 552

Text #17

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1475. Min Act: -0.0002

Data Index: 1000277 (C4 (Web Text))

Max Activating Token Index: 736

Text #18

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1482. Min Act: -0.0002

Data Index: 117395 (C4 (Web Text))

Max Activating Token Index: 105

Text #19

Max Range: 0.1730. Min Range: -0.1730

Max Act: 0.1421. Min Act: -0.0002

Data Index: 534124 (C4 (Web Text))

Max Activating Token Index: 385