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

Dataset: The Pile

Neuron 2314 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.0818. Min Range: -7.0818

Max Act: 7.0818. Min Act: -0.0000

Data Index: 1573068 (The Pile)

Max Activating Token Index: 494

Text #1

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.8022. Min Act: -0.0000

Data Index: 1887628 (The Pile)

Max Activating Token Index: 503

Text #2

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.5470. Min Act: -0.0000

Data Index: 1518990 (The Pile)

Max Activating Token Index: 1010

Text #3

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.7415. Min Act: -0.0000

Data Index: 776523 (The Pile)

Max Activating Token Index: 576

Text #4

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.8259. Min Act: -0.0000

Data Index: 1897923 (The Pile)

Max Activating Token Index: 267

Text #5

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.4009. Min Act: -0.0000

Data Index: 1345485 (The Pile)

Max Activating Token Index: 411

Text #6

Max Range: 7.0818. Min Range: -7.0818

Max Act: 5.9834. Min Act: -0.0000

Data Index: 1237787 (The Pile)

Max Activating Token Index: 874

Text #7

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.4909. Min Act: -0.0000

Data Index: 191917 (The Pile)

Max Activating Token Index: 800

Text #8

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.7269. Min Act: -0.0000

Data Index: 1074003 (The Pile)

Max Activating Token Index: 987

Text #9

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.4203. Min Act: -0.0000

Data Index: 109530 (The Pile)

Max Activating Token Index: 279

Text #10

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.3836. Min Act: -0.0000

Data Index: 1311145 (The Pile)

Max Activating Token Index: 850

Text #11

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.5406. Min Act: -0.0000

Data Index: 326386 (The Pile)

Max Activating Token Index: 886

Text #12

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.2973. Min Act: -0.0000

Data Index: 1670124 (The Pile)

Max Activating Token Index: 936

Text #13

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.4364. Min Act: -0.0000

Data Index: 275314 (The Pile)

Max Activating Token Index: 784

Text #14

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.0852. Min Act: -0.0000

Data Index: 582640 (The Pile)

Max Activating Token Index: 891

Text #15

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.3199. Min Act: -0.0000

Data Index: 735652 (The Pile)

Max Activating Token Index: 885

Text #16

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.4610. Min Act: -0.0000

Data Index: 1591545 (The Pile)

Max Activating Token Index: 687

Text #17

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.6405. Min Act: -0.0000

Data Index: 1426091 (The Pile)

Max Activating Token Index: 635

Text #18

Max Range: 7.0818. Min Range: -7.0818

Max Act: 6.1962. Min Act: -0.0000

Data Index: 1534183 (The Pile)

Max Activating Token Index: 267

Text #19

Max Range: 7.0818. Min Range: -7.0818

Max Act: 5.8057. Min Act: -0.0000

Data Index: 1714833 (The Pile)

Max Activating Token Index: 779