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

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

Neuron 256 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: 2.3061. Min Range: -2.3061

Max Act: 2.3061. Min Act: -0.0000

Data Index: 263964 (C4 (Web Text))

Max Activating Token Index: 801

Text #1

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.9895. Min Act: -0.0000

Data Index: 993905 (C4 (Web Text))

Max Activating Token Index: 993

Text #2

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.7543. Min Act: -0.0000

Data Index: 697903 (C4 (Web Text))

Max Activating Token Index: 359

Text #3

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.6119. Min Act: 0.0000

Data Index: 274652 (C4 (Web Text))

Max Activating Token Index: 49

Text #4

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.5573. Min Act: -0.0000

Data Index: 841367 (C4 (Web Text))

Max Activating Token Index: 145

Text #5

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.3455. Min Act: 0.0000

Data Index: 772773 (C4 (Web Text))

Max Activating Token Index: 561

Text #6

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.3366. Min Act: -0.0000

Data Index: 727141 (C4 (Web Text))

Max Activating Token Index: 33

Text #7

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.3290. Min Act: -0.0000

Data Index: 471879 (C4 (Web Text))

Max Activating Token Index: 990

Text #8

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.2981. Min Act: 0.0000

Data Index: 801020 (C4 (Web Text))

Max Activating Token Index: 20

Text #9

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.3684. Min Act: -0.0000

Data Index: 801458 (C4 (Web Text))

Max Activating Token Index: 326

Text #10

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.3583. Min Act: -0.0000

Data Index: 889641 (C4 (Web Text))

Max Activating Token Index: 931

Text #11

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.1792. Min Act: 0.0000

Data Index: 732617 (C4 (Web Text))

Max Activating Token Index: 92

Text #12

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.3162. Min Act: -0.0000

Data Index: 1201086 (C4 (Web Text))

Max Activating Token Index: 999

Text #13

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.2539. Min Act: 0.0000

Data Index: 1197605 (C4 (Web Text))

Max Activating Token Index: 460

Text #14

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.2377. Min Act: 0.0001

Data Index: 985407 (C4 (Web Text))

Max Activating Token Index: 923

Text #15

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.1320. Min Act: 0.0000

Data Index: 116832 (C4 (Web Text))

Max Activating Token Index: 82

Text #16

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.1937. Min Act: 0.0001

Data Index: 1397799 (Python Code)

Max Activating Token Index: 147

Text #17

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.1509. Min Act: -0.0000

Data Index: 20793 (C4 (Web Text))

Max Activating Token Index: 135

Text #18

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.1448. Min Act: -0.0000

Data Index: 297928 (C4 (Web Text))

Max Activating Token Index: 872

Text #19

Max Range: 2.3061. Min Range: -2.3061

Max Act: 1.1194. Min Act: -0.0000

Data Index: 448744 (C4 (Web Text))

Max Activating Token Index: 165