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

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

Neuron 3813 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.6865. Min Range: -2.6865

Max Act: 2.6865. Min Act: -0.0000

Data Index: 1525225 (Python Code)

Max Activating Token Index: 829

Text #1

Max Range: 2.6865. Min Range: -2.6865

Max Act: 2.3117. Min Act: -0.0000

Data Index: 595419 (C4 (Web Text))

Max Activating Token Index: 205

Text #2

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.7376. Min Act: -0.0000

Data Index: 1583605 (Python Code)

Max Activating Token Index: 966

Text #3

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.6669. Min Act: -0.0000

Data Index: 359262 (C4 (Web Text))

Max Activating Token Index: 724

Text #4

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.6796. Min Act: -0.0000

Data Index: 167497 (C4 (Web Text))

Max Activating Token Index: 287

Text #5

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.6152. Min Act: -0.0000

Data Index: 421567 (C4 (Web Text))

Max Activating Token Index: 493

Text #6

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.5010. Min Act: -0.0000

Data Index: 1549988 (Python Code)

Max Activating Token Index: 1022

Text #7

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.4743. Min Act: -0.0000

Data Index: 1505945 (Python Code)

Max Activating Token Index: 674

Text #8

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.2306. Min Act: -0.0000

Data Index: 342717 (C4 (Web Text))

Max Activating Token Index: 321

Text #9

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.5065. Min Act: -0.0000

Data Index: 991492 (C4 (Web Text))

Max Activating Token Index: 401

Text #10

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.5170. Min Act: -0.0000

Data Index: 391117 (C4 (Web Text))

Max Activating Token Index: 42

Text #11

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.4112. Min Act: -0.0000

Data Index: 1500604 (Python Code)

Max Activating Token Index: 607

Text #12

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.4406. Min Act: -0.0000

Data Index: 1311526 (C4 (Web Text))

Max Activating Token Index: 787

Text #13

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.7072. Min Act: -0.0000

Data Index: 17310 (C4 (Web Text))

Max Activating Token Index: 267

Text #14

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.3408. Min Act: -0.0000

Data Index: 1444887 (Python Code)

Max Activating Token Index: 226

Text #15

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.3304. Min Act: -0.0000

Data Index: 865476 (C4 (Web Text))

Max Activating Token Index: 426

Text #16

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.4168. Min Act: -0.0000

Data Index: 477079 (C4 (Web Text))

Max Activating Token Index: 939

Text #17

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.3944. Min Act: -0.0000

Data Index: 147039 (C4 (Web Text))

Max Activating Token Index: 47

Text #18

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.5939. Min Act: -0.0000

Data Index: 1010609 (C4 (Web Text))

Max Activating Token Index: 265

Text #19

Max Range: 2.6865. Min Range: -2.6865

Max Act: 1.4568. Min Act: -0.0000

Data Index: 174223 (C4 (Web Text))

Max Activating Token Index: 792