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

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

Neuron 3466 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: 4.2316. Min Range: -4.2316

Max Act: 4.2316. Min Act: -0.0000

Data Index: 4211 (C4 (Web Text))

Max Activating Token Index: 878

Text #1

Max Range: 4.2316. Min Range: -4.2316

Max Act: 3.4838. Min Act: -0.0000

Data Index: 617216 (C4 (Web Text))

Max Activating Token Index: 697

Text #2

Max Range: 4.2316. Min Range: -4.2316

Max Act: 3.6205. Min Act: -0.0000

Data Index: 351217 (C4 (Web Text))

Max Activating Token Index: 92

Text #3

Max Range: 4.2316. Min Range: -4.2316

Max Act: 3.0029. Min Act: -0.0000

Data Index: 1457385 (Python Code)

Max Activating Token Index: 415

Text #4

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.9725. Min Act: 0.0000

Data Index: 766113 (C4 (Web Text))

Max Activating Token Index: 420

Text #5

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.8480. Min Act: -0.0000

Data Index: 1030732 (C4 (Web Text))

Max Activating Token Index: 769

Text #6

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.7969. Min Act: 0.0000

Data Index: 428253 (C4 (Web Text))

Max Activating Token Index: 786

Text #7

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.6728. Min Act: -0.0000

Data Index: 387786 (C4 (Web Text))

Max Activating Token Index: 323

Text #8

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.6444. Min Act: -0.0000

Data Index: 447636 (C4 (Web Text))

Max Activating Token Index: 46

Text #9

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.5467. Min Act: -0.0000

Data Index: 1181062 (C4 (Web Text))

Max Activating Token Index: 188

Text #10

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.5256. Min Act: -0.0000

Data Index: 206564 (C4 (Web Text))

Max Activating Token Index: 275

Text #11

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.6303. Min Act: 0.0000

Data Index: 903857 (C4 (Web Text))

Max Activating Token Index: 696

Text #12

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.4915. Min Act: -0.0000

Data Index: 108786 (C4 (Web Text))

Max Activating Token Index: 917

Text #13

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.6097. Min Act: -0.0000

Data Index: 475428 (C4 (Web Text))

Max Activating Token Index: 555

Text #14

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.3428. Min Act: -0.0000

Data Index: 709775 (C4 (Web Text))

Max Activating Token Index: 241

Text #15

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.4237. Min Act: -0.0000

Data Index: 1446301 (Python Code)

Max Activating Token Index: 179

Text #16

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.5148. Min Act: 0.0000

Data Index: 1197351 (C4 (Web Text))

Max Activating Token Index: 45

Text #17

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.3911. Min Act: -0.0000

Data Index: 992401 (C4 (Web Text))

Max Activating Token Index: 807

Text #18

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.3132. Min Act: -0.0000

Data Index: 1186517 (C4 (Web Text))

Max Activating Token Index: 137

Text #19

Max Range: 4.2316. Min Range: -4.2316

Max Act: 2.3250. Min Act: -0.0000

Data Index: 1040115 (C4 (Web Text))

Max Activating Token Index: 542