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

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

Neuron 136 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: 0.5972. Min Range: -0.5972

Max Act: 0.5972. Min Act: -0.0001

Data Index: 453325 (C4 (Web Text))

Max Activating Token Index: 552

Text #1

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.5420. Min Act: -0.0001

Data Index: 466119 (C4 (Web Text))

Max Activating Token Index: 74

Text #2

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.5155. Min Act: -0.0001

Data Index: 370829 (C4 (Web Text))

Max Activating Token Index: 703

Text #3

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.5125. Min Act: -0.0001

Data Index: 420433 (C4 (Web Text))

Max Activating Token Index: 671

Text #4

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.5104. Min Act: -0.0001

Data Index: 1158867 (C4 (Web Text))

Max Activating Token Index: 625

Text #5

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.5114. Min Act: -0.0001

Data Index: 47110 (C4 (Web Text))

Max Activating Token Index: 372

Text #6

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.5049. Min Act: -0.0001

Data Index: 409976 (C4 (Web Text))

Max Activating Token Index: 592

Text #7

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.5036. Min Act: -0.0001

Data Index: 1128642 (C4 (Web Text))

Max Activating Token Index: 416

Text #8

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.5039. Min Act: -0.0001

Data Index: 480623 (C4 (Web Text))

Max Activating Token Index: 673

Text #9

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4941. Min Act: -0.0001

Data Index: 7221 (C4 (Web Text))

Max Activating Token Index: 442

Text #10

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4936. Min Act: -0.0001

Data Index: 984829 (C4 (Web Text))

Max Activating Token Index: 697

Text #11

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4874. Min Act: -0.0001

Data Index: 305997 (C4 (Web Text))

Max Activating Token Index: 405

Text #12

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4840. Min Act: -0.0001

Data Index: 1022616 (C4 (Web Text))

Max Activating Token Index: 906

Text #13

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4561. Min Act: -0.0001

Data Index: 1056500 (C4 (Web Text))

Max Activating Token Index: 775

Text #14

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4569. Min Act: -0.0001

Data Index: 623378 (C4 (Web Text))

Max Activating Token Index: 521

Text #15

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4571. Min Act: -0.0001

Data Index: 590575 (C4 (Web Text))

Max Activating Token Index: 311

Text #16

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4691. Min Act: -0.0001

Data Index: 1002372 (C4 (Web Text))

Max Activating Token Index: 267

Text #17

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4672. Min Act: -0.0001

Data Index: 1092469 (C4 (Web Text))

Max Activating Token Index: 47

Text #18

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4745. Min Act: -0.0001

Data Index: 1312178 (C4 (Web Text))

Max Activating Token Index: 615

Text #19

Max Range: 0.5972. Min Range: -0.5972

Max Act: 0.4757. Min Act: -0.0001

Data Index: 1265514 (C4 (Web Text))

Max Activating Token Index: 168