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

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

Neuron 356 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1856. Min Range: -1.1856

Max Act: 1.1045. Min Act: -0.0001

Data Index: 381017 (C4 (Web Text))

Max Activating Token Index: 309

Text #1

Max Range: 1.1856. Min Range: -1.1856

Max Act: 1.1856. Min Act: -0.0001

Data Index: 386983 (C4 (Web Text))

Max Activating Token Index: 72

Text #2

Max Range: 1.1856. Min Range: -1.1856

Max Act: 1.0777. Min Act: -0.0001

Data Index: 772138 (C4 (Web Text))

Max Activating Token Index: 106

Text #3

Max Range: 1.1856. Min Range: -1.1856

Max Act: 1.0717. Min Act: -0.0001

Data Index: 279131 (C4 (Web Text))

Max Activating Token Index: 553

Text #4

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8841. Min Act: -0.0001

Data Index: 1060398 (C4 (Web Text))

Max Activating Token Index: 65

Text #5

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.9479. Min Act: -0.0001

Data Index: 46864 (C4 (Web Text))

Max Activating Token Index: 550

Text #6

Max Range: 1.1856. Min Range: -1.1856

Max Act: 1.0130. Min Act: -0.0001

Data Index: 457982 (C4 (Web Text))

Max Activating Token Index: 225

Text #7

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8751. Min Act: -0.0001

Data Index: 522223 (C4 (Web Text))

Max Activating Token Index: 791

Text #8

Max Range: 1.1856. Min Range: -1.1856

Max Act: 1.0245. Min Act: -0.0001

Data Index: 496831 (C4 (Web Text))

Max Activating Token Index: 852

Text #9

Max Range: 1.1856. Min Range: -1.1856

Max Act: 1.0181. Min Act: -0.0001

Data Index: 24173 (C4 (Web Text))

Max Activating Token Index: 666

Text #10

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8539. Min Act: -0.0001

Data Index: 728427 (C4 (Web Text))

Max Activating Token Index: 848

Text #11

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8394. Min Act: -0.0001

Data Index: 1443229 (Python Code)

Max Activating Token Index: 264

Text #12

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8376. Min Act: -0.0001

Data Index: 1455030 (Python Code)

Max Activating Token Index: 8

Text #13

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8660. Min Act: -0.0001

Data Index: 287476 (C4 (Web Text))

Max Activating Token Index: 628

Text #14

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8713. Min Act: -0.0001

Data Index: 1334265 (C4 (Web Text))

Max Activating Token Index: 62

Text #15

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8566. Min Act: -0.0001

Data Index: 705135 (C4 (Web Text))

Max Activating Token Index: 429

Text #16

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.7975. Min Act: -0.0001

Data Index: 1345037 (C4 (Web Text))

Max Activating Token Index: 41

Text #17

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.7593. Min Act: -0.0001

Data Index: 315537 (C4 (Web Text))

Max Activating Token Index: 767

Text #18

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8379. Min Act: -0.0001

Data Index: 554632 (C4 (Web Text))

Max Activating Token Index: 248

Text #19

Max Range: 1.1856. Min Range: -1.1856

Max Act: 0.8320. Min Act: -0.0001

Data Index: 1429068 (Python Code)

Max Activating Token Index: 954