Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4772 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.6950. Min Act: -0.0000

Data Index: 952208 (C4 (Web Text))

Max Activating Token Index: 252

Text #1

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.5316. Min Act: -0.0000

Data Index: 201550 (C4 (Web Text))

Max Activating Token Index: 818

Text #2

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.2523. Min Act: -0.0000

Data Index: 1076917 (C4 (Web Text))

Max Activating Token Index: 847

Text #3

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.2331. Min Act: -0.0000

Data Index: 828770 (C4 (Web Text))

Max Activating Token Index: 468

Text #4

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.2359. Min Act: -0.0000

Data Index: 1548737 (Python Code)

Max Activating Token Index: 247

Text #5

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.2469. Min Act: -0.0000

Data Index: 383840 (C4 (Web Text))

Max Activating Token Index: 1016

Text #6

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.2025. Min Act: -0.0000

Data Index: 676347 (C4 (Web Text))

Max Activating Token Index: 954

Text #7

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.3313. Min Act: -0.0000

Data Index: 673284 (C4 (Web Text))

Max Activating Token Index: 47

Text #8

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.1638. Min Act: -0.0000

Data Index: 486071 (C4 (Web Text))

Max Activating Token Index: 315

Text #9

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.1606. Min Act: -0.0000

Data Index: 883089 (C4 (Web Text))

Max Activating Token Index: 239

Text #10

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.1487. Min Act: -0.0000

Data Index: 130094 (C4 (Web Text))

Max Activating Token Index: 611

Text #11

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.0020. Min Act: -0.0000

Data Index: 1169813 (C4 (Web Text))

Max Activating Token Index: 308

Text #12

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.1226. Min Act: -0.0000

Data Index: 422135 (C4 (Web Text))

Max Activating Token Index: 1019

Text #13

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.0960. Min Act: -0.0000

Data Index: 132414 (C4 (Web Text))

Max Activating Token Index: 43

Text #14

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.0930. Min Act: -0.0000

Data Index: 1165211 (C4 (Web Text))

Max Activating Token Index: 256

Text #15

Max Range: 2.6950. Min Range: -2.6950

Max Act: 1.9874. Min Act: -0.0000

Data Index: 254088 (C4 (Web Text))

Max Activating Token Index: 839

Text #16

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.0660. Min Act: -0.0000

Data Index: 218343 (C4 (Web Text))

Max Activating Token Index: 160

Text #17

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.0715. Min Act: -0.0000

Data Index: 41567 (C4 (Web Text))

Max Activating Token Index: 26

Text #18

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.0834. Min Act: -0.0000

Data Index: 801082 (C4 (Web Text))

Max Activating Token Index: 596

Text #19

Max Range: 2.6950. Min Range: -2.6950

Max Act: 2.1797. Min Act: -0.0000

Data Index: 341169 (C4 (Web Text))

Max Activating Token Index: 790