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

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

Neuron 2005 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: 2.8700. Min Range: -2.8700

Max Act: 2.8700. Min Act: -0.0000

Data Index: 758331 (C4 (Web Text))

Max Activating Token Index: 366

Text #1

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.8486. Min Act: -0.0000

Data Index: 1461948 (Python Code)

Max Activating Token Index: 461

Text #2

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.7764. Min Act: -0.0000

Data Index: 678035 (C4 (Web Text))

Max Activating Token Index: 540

Text #3

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.7118. Min Act: -0.0000

Data Index: 1529017 (Python Code)

Max Activating Token Index: 675

Text #4

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.7429. Min Act: -0.0000

Data Index: 363947 (C4 (Web Text))

Max Activating Token Index: 296

Text #5

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.8062. Min Act: -0.0000

Data Index: 1538985 (Python Code)

Max Activating Token Index: 941

Text #6

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.7345. Min Act: -0.0000

Data Index: 1530009 (Python Code)

Max Activating Token Index: 200

Text #7

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.7218. Min Act: -0.0000

Data Index: 1454897 (Python Code)

Max Activating Token Index: 1009

Text #8

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.7049. Min Act: -0.0000

Data Index: 1552773 (Python Code)

Max Activating Token Index: 542

Text #9

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.6742. Min Act: -0.0000

Data Index: 1621907 (Python Code)

Max Activating Token Index: 355

Text #10

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.7630. Min Act: -0.0000

Data Index: 133791 (C4 (Web Text))

Max Activating Token Index: 422

Text #11

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.5306. Min Act: -0.0000

Data Index: 1229996 (C4 (Web Text))

Max Activating Token Index: 153

Text #12

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.5735. Min Act: -0.0000

Data Index: 1131493 (C4 (Web Text))

Max Activating Token Index: 257

Text #13

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.5185. Min Act: -0.0000

Data Index: 1530296 (Python Code)

Max Activating Token Index: 420

Text #14

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.4433. Min Act: -0.0000

Data Index: 456064 (C4 (Web Text))

Max Activating Token Index: 642

Text #15

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.4259. Min Act: -0.0000

Data Index: 1506718 (Python Code)

Max Activating Token Index: 20

Text #16

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.3994. Min Act: -0.0000

Data Index: 1609118 (Python Code)

Max Activating Token Index: 310

Text #17

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.4765. Min Act: -0.0000

Data Index: 394294 (C4 (Web Text))

Max Activating Token Index: 141

Text #18

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.4762. Min Act: -0.0000

Data Index: 1100170 (C4 (Web Text))

Max Activating Token Index: 219

Text #19

Max Range: 2.8700. Min Range: -2.8700

Max Act: 1.3828. Min Act: -0.0000

Data Index: 1523376 (Python Code)

Max Activating Token Index: 479