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

Dataset: The Pile

Neuron 3737 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: 10.9916. Min Range: -10.9916

Max Act: 10.9916. Min Act: -0.0000

Data Index: 189495 (The Pile)

Max Activating Token Index: 114

Text #1

Max Range: 10.9916. Min Range: -10.9916

Max Act: 10.3165. Min Act: -0.0000

Data Index: 1350303 (The Pile)

Max Activating Token Index: 336

Text #2

Max Range: 10.9916. Min Range: -10.9916

Max Act: 10.3698. Min Act: -0.0000

Data Index: 550331 (The Pile)

Max Activating Token Index: 69

Text #3

Max Range: 10.9916. Min Range: -10.9916

Max Act: 10.2044. Min Act: -0.0000

Data Index: 1181206 (The Pile)

Max Activating Token Index: 262

Text #4

Max Range: 10.9916. Min Range: -10.9916

Max Act: 10.3010. Min Act: -0.0000

Data Index: 1501519 (The Pile)

Max Activating Token Index: 308

Text #5

Max Range: 10.9916. Min Range: -10.9916

Max Act: 10.0174. Min Act: -0.0000

Data Index: 846188 (The Pile)

Max Activating Token Index: 124

Text #6

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.3210. Min Act: -0.0000

Data Index: 181391 (The Pile)

Max Activating Token Index: 89

Text #7

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.9383. Min Act: -0.0000

Data Index: 1262244 (The Pile)

Max Activating Token Index: 54

Text #8

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.8497. Min Act: -0.0000

Data Index: 1601829 (The Pile)

Max Activating Token Index: 308

Text #9

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.8747. Min Act: -0.0000

Data Index: 323526 (The Pile)

Max Activating Token Index: 172

Text #10

Max Range: 10.9916. Min Range: -10.9916

Max Act: 10.0513. Min Act: -0.0000

Data Index: 841421 (The Pile)

Max Activating Token Index: 281

Text #11

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.9345. Min Act: -0.0000

Data Index: 113178 (The Pile)

Max Activating Token Index: 485

Text #12

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.8737. Min Act: -0.0000

Data Index: 787416 (The Pile)

Max Activating Token Index: 132

Text #13

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.8588. Min Act: -0.0000

Data Index: 1506119 (The Pile)

Max Activating Token Index: 963

Text #14

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.7741. Min Act: -0.0000

Data Index: 378998 (The Pile)

Max Activating Token Index: 968

Text #15

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.9118. Min Act: -0.0000

Data Index: 1330972 (The Pile)

Max Activating Token Index: 266

Text #16

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.4198. Min Act: -0.0000

Data Index: 1976244 (The Pile)

Max Activating Token Index: 441

Text #17

Max Range: 10.9916. Min Range: -10.9916

Max Act: 10.0297. Min Act: -0.0000

Data Index: 1742050 (The Pile)

Max Activating Token Index: 65

Text #18

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.6006. Min Act: -0.0000

Data Index: 725671 (The Pile)

Max Activating Token Index: 74

Text #19

Max Range: 10.9916. Min Range: -10.9916

Max Act: 9.8400. Min Act: -0.0000

Data Index: 1328052 (The Pile)

Max Activating Token Index: 329