Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 256 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3954. Min Act: -0.0001

Data Index: 881659 (C4 (Web Text))

Max Activating Token Index: 933

Text #1

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3880. Min Act: -0.0001

Data Index: 939313 (C4 (Web Text))

Max Activating Token Index: 501

Text #2

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.4123. Min Act: -0.0001

Data Index: 1337487 (C4 (Web Text))

Max Activating Token Index: 171

Text #3

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.4043. Min Act: -0.0001

Data Index: 560317 (C4 (Web Text))

Max Activating Token Index: 91

Text #4

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3802. Min Act: -0.0001

Data Index: 196749 (C4 (Web Text))

Max Activating Token Index: 135

Text #5

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3630. Min Act: -0.0001

Data Index: 997691 (C4 (Web Text))

Max Activating Token Index: 627

Text #6

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3989. Min Act: -0.0001

Data Index: 356839 (C4 (Web Text))

Max Activating Token Index: 635

Text #7

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3596. Min Act: -0.0001

Data Index: 57894 (C4 (Web Text))

Max Activating Token Index: 632

Text #8

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3537. Min Act: -0.0001

Data Index: 1288460 (C4 (Web Text))

Max Activating Token Index: 780

Text #9

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3730. Min Act: -0.0001

Data Index: 1725 (C4 (Web Text))

Max Activating Token Index: 758

Text #10

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3531. Min Act: -0.0001

Data Index: 373988 (C4 (Web Text))

Max Activating Token Index: 851

Text #11

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3543. Min Act: -0.0001

Data Index: 713110 (C4 (Web Text))

Max Activating Token Index: 533

Text #12

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3216. Min Act: -0.0001

Data Index: 1065947 (C4 (Web Text))

Max Activating Token Index: 175

Text #13

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3389. Min Act: -0.0001

Data Index: 1099245 (C4 (Web Text))

Max Activating Token Index: 958

Text #14

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3620. Min Act: -0.0001

Data Index: 294150 (C4 (Web Text))

Max Activating Token Index: 861

Text #15

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3223. Min Act: -0.0001

Data Index: 430480 (C4 (Web Text))

Max Activating Token Index: 30

Text #16

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3465. Min Act: -0.0001

Data Index: 499187 (C4 (Web Text))

Max Activating Token Index: 675

Text #17

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3354. Min Act: -0.0001

Data Index: 546823 (C4 (Web Text))

Max Activating Token Index: 332

Text #18

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3191. Min Act: -0.0001

Data Index: 780561 (C4 (Web Text))

Max Activating Token Index: 390

Text #19

Max Range: 0.4123. Min Range: -0.4123

Max Act: 0.3115. Min Act: -0.0001

Data Index: 922384 (C4 (Web Text))

Max Activating Token Index: 415