Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2746 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 3.5673. Min Range: -3.5673

Max Act: 3.5673. Min Act: -0.0001

Data Index: 968434 (C4 (Web Text))

Max Activating Token Index: 395

Text #1

Max Range: 3.5673. Min Range: -3.5673

Max Act: 3.4422. Min Act: -0.0000

Data Index: 857679 (C4 (Web Text))

Max Activating Token Index: 238

Text #2

Max Range: 3.5673. Min Range: -3.5673

Max Act: 3.2298. Min Act: -0.0001

Data Index: 375920 (C4 (Web Text))

Max Activating Token Index: 537

Text #3

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.9531. Min Act: -0.0001

Data Index: 471642 (C4 (Web Text))

Max Activating Token Index: 328

Text #4

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.6360. Min Act: -0.0000

Data Index: 489689 (C4 (Web Text))

Max Activating Token Index: 570

Text #5

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.5768. Min Act: -0.0001

Data Index: 573063 (C4 (Web Text))

Max Activating Token Index: 725

Text #6

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.5443. Min Act: -0.0000

Data Index: 356995 (C4 (Web Text))

Max Activating Token Index: 101

Text #7

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.4966. Min Act: -0.0000

Data Index: 502272 (C4 (Web Text))

Max Activating Token Index: 115

Text #8

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.4420. Min Act: -0.0001

Data Index: 360992 (C4 (Web Text))

Max Activating Token Index: 627

Text #9

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.3177. Min Act: -0.0000

Data Index: 826559 (C4 (Web Text))

Max Activating Token Index: 905

Text #10

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.4349. Min Act: -0.0001

Data Index: 1577992 (Python Code)

Max Activating Token Index: 615

Text #11

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.3951. Min Act: -0.0000

Data Index: 1286212 (C4 (Web Text))

Max Activating Token Index: 439

Text #12

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.2835. Min Act: -0.0000

Data Index: 668894 (C4 (Web Text))

Max Activating Token Index: 814

Text #13

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.3197. Min Act: -0.0000

Data Index: 1292638 (C4 (Web Text))

Max Activating Token Index: 741

Text #14

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.3713. Min Act: -0.0001

Data Index: 1143717 (C4 (Web Text))

Max Activating Token Index: 148

Text #15

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.4796. Min Act: -0.0001

Data Index: 334806 (C4 (Web Text))

Max Activating Token Index: 965

Text #16

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.1666. Min Act: -0.0000

Data Index: 140088 (C4 (Web Text))

Max Activating Token Index: 709

Text #17

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.3290. Min Act: -0.0000

Data Index: 1589488 (Python Code)

Max Activating Token Index: 987

Text #18

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.3033. Min Act: -0.0000

Data Index: 666987 (C4 (Web Text))

Max Activating Token Index: 422

Text #19

Max Range: 3.5673. Min Range: -3.5673

Max Act: 2.2846. Min Act: -0.0000

Data Index: 1041733 (C4 (Web Text))

Max Activating Token Index: 317