Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 794 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.3997. Min Range: -2.3997

Max Act: 2.3997. Min Act: -0.0000

Data Index: 157432 (C4 (Web Text))

Max Activating Token Index: 416

Text #1

Max Range: 2.3997. Min Range: -2.3997

Max Act: 2.2756. Min Act: -0.0000

Data Index: 1167203 (C4 (Web Text))

Max Activating Token Index: 323

Text #2

Max Range: 2.3997. Min Range: -2.3997

Max Act: 2.2147. Min Act: -0.0000

Data Index: 308724 (C4 (Web Text))

Max Activating Token Index: 773

Text #3

Max Range: 2.3997. Min Range: -2.3997

Max Act: 2.0495. Min Act: -0.0000

Data Index: 90582 (C4 (Web Text))

Max Activating Token Index: 430

Text #4

Max Range: 2.3997. Min Range: -2.3997

Max Act: 2.0050. Min Act: -0.0000

Data Index: 1627631 (Python Code)

Max Activating Token Index: 96

Text #5

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.9205. Min Act: -0.0000

Data Index: 599402 (C4 (Web Text))

Max Activating Token Index: 140

Text #6

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.8884. Min Act: -0.0000

Data Index: 711715 (C4 (Web Text))

Max Activating Token Index: 741

Text #7

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.8231. Min Act: -0.0000

Data Index: 933960 (C4 (Web Text))

Max Activating Token Index: 464

Text #8

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.9270. Min Act: -0.0000

Data Index: 323042 (C4 (Web Text))

Max Activating Token Index: 595

Text #9

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.8074. Min Act: -0.0000

Data Index: 1647280 (Python Code)

Max Activating Token Index: 264

Text #10

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.6784. Min Act: -0.0000

Data Index: 1381839 (Python Code)

Max Activating Token Index: 214

Text #11

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.6967. Min Act: -0.0000

Data Index: 1210793 (C4 (Web Text))

Max Activating Token Index: 697

Text #12

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.6724. Min Act: -0.0000

Data Index: 1360973 (Python Code)

Max Activating Token Index: 194

Text #13

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.6616. Min Act: -0.0000

Data Index: 1345670 (C4 (Web Text))

Max Activating Token Index: 618

Text #14

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.6341. Min Act: -0.0000

Data Index: 630660 (C4 (Web Text))

Max Activating Token Index: 400

Text #15

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.5420. Min Act: -0.0000

Data Index: 180118 (C4 (Web Text))

Max Activating Token Index: 569

Text #16

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.7048. Min Act: -0.0000

Data Index: 917081 (C4 (Web Text))

Max Activating Token Index: 679

Text #17

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.5229. Min Act: -0.0000

Data Index: 853056 (C4 (Web Text))

Max Activating Token Index: 172

Text #18

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.5866. Min Act: -0.0000

Data Index: 1294055 (C4 (Web Text))

Max Activating Token Index: 472

Text #19

Max Range: 2.3997. Min Range: -2.3997

Max Act: 1.5819. Min Act: -0.0000

Data Index: 918879 (C4 (Web Text))

Max Activating Token Index: 518