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

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

Neuron 4708 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.6657. Min Act: 0.0000

Data Index: 1158981 (C4 (Web Text))

Max Activating Token Index: 245

Text #1

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.6610. Min Act: -0.0000

Data Index: 1063791 (C4 (Web Text))

Max Activating Token Index: 29

Text #2

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.5443. Min Act: -0.0000

Data Index: 1486092 (Python Code)

Max Activating Token Index: 374

Text #3

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.5022. Min Act: -0.0001

Data Index: 1154234 (C4 (Web Text))

Max Activating Token Index: 322

Text #4

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.3327. Min Act: -0.0000

Data Index: 899141 (C4 (Web Text))

Max Activating Token Index: 852

Text #5

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.3106. Min Act: -0.0000

Data Index: 266329 (C4 (Web Text))

Max Activating Token Index: 9

Text #6

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.5087. Min Act: -0.0000

Data Index: 490018 (C4 (Web Text))

Max Activating Token Index: 584

Text #7

Max Range: 2.6657. Min Range: -2.6657

Max Act: 1.9610. Min Act: -0.0000

Data Index: 599168 (C4 (Web Text))

Max Activating Token Index: 528

Text #8

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.0176. Min Act: -0.0000

Data Index: 393548 (C4 (Web Text))

Max Activating Token Index: 200

Text #9

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.0885. Min Act: -0.0000

Data Index: 208345 (C4 (Web Text))

Max Activating Token Index: 78

Text #10

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.0608. Min Act: -0.0000

Data Index: 900622 (C4 (Web Text))

Max Activating Token Index: 58

Text #11

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.0614. Min Act: -0.0000

Data Index: 756219 (C4 (Web Text))

Max Activating Token Index: 72

Text #12

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.1448. Min Act: -0.0000

Data Index: 1027072 (C4 (Web Text))

Max Activating Token Index: 478

Text #13

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.0012. Min Act: -0.0000

Data Index: 791556 (C4 (Web Text))

Max Activating Token Index: 193

Text #14

Max Range: 2.6657. Min Range: -2.6657

Max Act: 1.9867. Min Act: -0.0000

Data Index: 485303 (C4 (Web Text))

Max Activating Token Index: 866

Text #15

Max Range: 2.6657. Min Range: -2.6657

Max Act: 2.0924. Min Act: -0.0000

Data Index: 1274854 (C4 (Web Text))

Max Activating Token Index: 57

Text #16

Max Range: 2.6657. Min Range: -2.6657

Max Act: 1.9574. Min Act: -0.0000

Data Index: 1322177 (C4 (Web Text))

Max Activating Token Index: 32

Text #17

Max Range: 2.6657. Min Range: -2.6657

Max Act: 1.9809. Min Act: 0.0000

Data Index: 806276 (C4 (Web Text))

Max Activating Token Index: 1020

Text #18

Max Range: 2.6657. Min Range: -2.6657

Max Act: 1.9479. Min Act: 0.0000

Data Index: 1128473 (C4 (Web Text))

Max Activating Token Index: 103

Text #19

Max Range: 2.6657. Min Range: -2.6657

Max Act: 1.9008. Min Act: -0.0000

Data Index: 48100 (C4 (Web Text))

Max Activating Token Index: 21