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

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

Neuron 1199 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.9574. Min Act: -0.0001

Data Index: 670059 (C4 (Web Text))

Max Activating Token Index: 118

Text #1

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.9636. Min Act: -0.0001

Data Index: 1156099 (C4 (Web Text))

Max Activating Token Index: 335

Text #2

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.9271. Min Act: -0.0001

Data Index: 1298548 (C4 (Web Text))

Max Activating Token Index: 518

Text #3

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.8828. Min Act: -0.0001

Data Index: 735183 (C4 (Web Text))

Max Activating Token Index: 81

Text #4

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.8424. Min Act: -0.0001

Data Index: 1159375 (C4 (Web Text))

Max Activating Token Index: 672

Text #5

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.8640. Min Act: -0.0001

Data Index: 556372 (C4 (Web Text))

Max Activating Token Index: 438

Text #6

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.8110. Min Act: -0.0001

Data Index: 1088359 (C4 (Web Text))

Max Activating Token Index: 456

Text #7

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.7224. Min Act: -0.0001

Data Index: 147184 (C4 (Web Text))

Max Activating Token Index: 541

Text #8

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.8019. Min Act: -0.0001

Data Index: 146329 (C4 (Web Text))

Max Activating Token Index: 635

Text #9

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.7974. Min Act: -0.0001

Data Index: 264335 (C4 (Web Text))

Max Activating Token Index: 662

Text #10

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.8261. Min Act: -0.0001

Data Index: 374659 (C4 (Web Text))

Max Activating Token Index: 120

Text #11

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.7661. Min Act: -0.0001

Data Index: 1533888 (Python Code)

Max Activating Token Index: 150

Text #12

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.7383. Min Act: -0.0001

Data Index: 1011172 (C4 (Web Text))

Max Activating Token Index: 287

Text #13

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.6939. Min Act: -0.0001

Data Index: 857562 (C4 (Web Text))

Max Activating Token Index: 971

Text #14

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.6446. Min Act: -0.0001

Data Index: 158174 (C4 (Web Text))

Max Activating Token Index: 223

Text #15

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.6768. Min Act: -0.0001

Data Index: 178217 (C4 (Web Text))

Max Activating Token Index: 524

Text #16

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.5391. Min Act: -0.0001

Data Index: 676765 (C4 (Web Text))

Max Activating Token Index: 246

Text #17

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.6726. Min Act: -0.0001

Data Index: 984436 (C4 (Web Text))

Max Activating Token Index: 246

Text #18

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.5485. Min Act: -0.0001

Data Index: 1411238 (Python Code)

Max Activating Token Index: 116

Text #19

Max Range: 0.9636. Min Range: -0.9636

Max Act: 0.7143. Min Act: -0.0001

Data Index: 887435 (C4 (Web Text))

Max Activating Token Index: 63