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

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

Neuron 2923 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.9188. Min Act: -0.0001

Data Index: 554821 (C4 (Web Text))

Max Activating Token Index: 851

Text #1

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.9621. Min Act: -0.0001

Data Index: 143314 (C4 (Web Text))

Max Activating Token Index: 874

Text #2

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8988. Min Act: -0.0001

Data Index: 906458 (C4 (Web Text))

Max Activating Token Index: 192

Text #3

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.9265. Min Act: -0.0001

Data Index: 607221 (C4 (Web Text))

Max Activating Token Index: 348

Text #4

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.9132. Min Act: -0.0001

Data Index: 146688 (C4 (Web Text))

Max Activating Token Index: 831

Text #5

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.9002. Min Act: -0.0001

Data Index: 1256192 (C4 (Web Text))

Max Activating Token Index: 698

Text #6

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8725. Min Act: -0.0001

Data Index: 388596 (C4 (Web Text))

Max Activating Token Index: 184

Text #7

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8660. Min Act: -0.0001

Data Index: 876747 (C4 (Web Text))

Max Activating Token Index: 304

Text #8

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8837. Min Act: -0.0001

Data Index: 1070598 (C4 (Web Text))

Max Activating Token Index: 94

Text #9

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8548. Min Act: -0.0001

Data Index: 493386 (C4 (Web Text))

Max Activating Token Index: 360

Text #10

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8564. Min Act: -0.0001

Data Index: 1656436 (Python Code)

Max Activating Token Index: 922

Text #11

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8395. Min Act: -0.0001

Data Index: 1074056 (C4 (Web Text))

Max Activating Token Index: 236

Text #12

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8048. Min Act: -0.0001

Data Index: 601689 (C4 (Web Text))

Max Activating Token Index: 632

Text #13

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.7997. Min Act: -0.0001

Data Index: 1222317 (C4 (Web Text))

Max Activating Token Index: 782

Text #14

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.7547. Min Act: -0.0001

Data Index: 1258412 (C4 (Web Text))

Max Activating Token Index: 101

Text #15

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.7363. Min Act: -0.0001

Data Index: 678840 (C4 (Web Text))

Max Activating Token Index: 574

Text #16

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8352. Min Act: -0.0001

Data Index: 378732 (C4 (Web Text))

Max Activating Token Index: 201

Text #17

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.8263. Min Act: -0.0001

Data Index: 718959 (C4 (Web Text))

Max Activating Token Index: 512

Text #18

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.7800. Min Act: -0.0001

Data Index: 619939 (C4 (Web Text))

Max Activating Token Index: 109

Text #19

Max Range: 0.9621. Min Range: -0.9621

Max Act: 0.7819. Min Act: -0.0001

Data Index: 1049326 (C4 (Web Text))

Max Activating Token Index: 642