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

Dataset: The Pile

Neuron 3248 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 10.7574. Min Range: -10.7574

Max Act: 10.7574. Min Act: -0.0000

Data Index: 842970 (The Pile)

Max Activating Token Index: 119

Text #1

Max Range: 10.7574. Min Range: -10.7574

Max Act: 10.6439. Min Act: -0.0000

Data Index: 233816 (The Pile)

Max Activating Token Index: 974

Text #2

Max Range: 10.7574. Min Range: -10.7574

Max Act: 10.7462. Min Act: -0.0000

Data Index: 1999723 (The Pile)

Max Activating Token Index: 488

Text #3

Max Range: 10.7574. Min Range: -10.7574

Max Act: 10.5275. Min Act: -0.0000

Data Index: 446798 (The Pile)

Max Activating Token Index: 272

Text #4

Max Range: 10.7574. Min Range: -10.7574

Max Act: 10.3672. Min Act: -0.0000

Data Index: 448533 (The Pile)

Max Activating Token Index: 91

Text #5

Max Range: 10.7574. Min Range: -10.7574

Max Act: 10.6024. Min Act: -0.0000

Data Index: 942494 (The Pile)

Max Activating Token Index: 262

Text #6

Max Range: 10.7574. Min Range: -10.7574

Max Act: 10.7415. Min Act: -0.0000

Data Index: 691942 (The Pile)

Max Activating Token Index: 573

Text #7

Max Range: 10.7574. Min Range: -10.7574

Max Act: 10.3351. Min Act: -0.0000

Data Index: 1410202 (The Pile)

Max Activating Token Index: 52

Text #8

Max Range: 10.7574. Min Range: -10.7574

Max Act: 10.0010. Min Act: -0.0000

Data Index: 1138882 (The Pile)

Max Activating Token Index: 244

Text #9

Max Range: 10.7574. Min Range: -10.7574

Max Act: 9.8526. Min Act: -0.0000

Data Index: 1939180 (The Pile)

Max Activating Token Index: 150

Text #10

Max Range: 10.7574. Min Range: -10.7574

Max Act: 9.4316. Min Act: -0.0000

Data Index: 15670 (The Pile)

Max Activating Token Index: 627

Text #11

Max Range: 10.7574. Min Range: -10.7574

Max Act: 9.9863. Min Act: -0.0000

Data Index: 64751 (The Pile)

Max Activating Token Index: 748

Text #12

Max Range: 10.7574. Min Range: -10.7574

Max Act: 8.9975. Min Act: -0.0000

Data Index: 326958 (The Pile)

Max Activating Token Index: 686

Text #13

Max Range: 10.7574. Min Range: -10.7574

Max Act: 9.6448. Min Act: -0.0000

Data Index: 234642 (The Pile)

Max Activating Token Index: 131

Text #14

Max Range: 10.7574. Min Range: -10.7574

Max Act: 9.6054. Min Act: -0.0000

Data Index: 853074 (The Pile)

Max Activating Token Index: 1009

Text #15

Max Range: 10.7574. Min Range: -10.7574

Max Act: 9.2415. Min Act: -0.0000

Data Index: 899770 (The Pile)

Max Activating Token Index: 707

Text #16

Max Range: 10.7574. Min Range: -10.7574

Max Act: 9.5972. Min Act: -0.0000

Data Index: 805499 (The Pile)

Max Activating Token Index: 524

Text #17

Max Range: 10.7574. Min Range: -10.7574

Max Act: 7.6555. Min Act: -0.0000

Data Index: 838648 (The Pile)

Max Activating Token Index: 387

Text #18

Max Range: 10.7574. Min Range: -10.7574

Max Act: 9.5459. Min Act: -0.0000

Data Index: 1042625 (The Pile)

Max Activating Token Index: 936

Text #19

Max Range: 10.7574. Min Range: -10.7574

Max Act: 7.9803. Min Act: -0.0000

Data Index: 1277631 (The Pile)

Max Activating Token Index: 973