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

Dataset: The Pile

Neuron 2885 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: 11.0219. Min Range: -11.0219

Max Act: 11.0219. Min Act: -0.0000

Data Index: 450279 (The Pile)

Max Activating Token Index: 325

Text #1

Max Range: 11.0219. Min Range: -11.0219

Max Act: 10.4350. Min Act: -0.0000

Data Index: 904465 (The Pile)

Max Activating Token Index: 535

Text #2

Max Range: 11.0219. Min Range: -11.0219

Max Act: 8.5628. Min Act: -0.0000

Data Index: 1095442 (The Pile)

Max Activating Token Index: 817

Text #3

Max Range: 11.0219. Min Range: -11.0219

Max Act: 8.3404. Min Act: 0.0001

Data Index: 663001 (The Pile)

Max Activating Token Index: 475

Text #4

Max Range: 11.0219. Min Range: -11.0219

Max Act: 8.2998. Min Act: -0.0000

Data Index: 541184 (The Pile)

Max Activating Token Index: 587

Text #5

Max Range: 11.0219. Min Range: -11.0219

Max Act: 8.2640. Min Act: -0.0000

Data Index: 1319458 (The Pile)

Max Activating Token Index: 367

Text #6

Max Range: 11.0219. Min Range: -11.0219

Max Act: 7.7929. Min Act: -0.0000

Data Index: 1667545 (The Pile)

Max Activating Token Index: 160

Text #7

Max Range: 11.0219. Min Range: -11.0219

Max Act: 7.0021. Min Act: -0.0000

Data Index: 336273 (The Pile)

Max Activating Token Index: 450

Text #8

Max Range: 11.0219. Min Range: -11.0219

Max Act: 7.1268. Min Act: -0.0000

Data Index: 1118934 (The Pile)

Max Activating Token Index: 206

Text #9

Max Range: 11.0219. Min Range: -11.0219

Max Act: 6.5181. Min Act: -0.0000

Data Index: 7989 (The Pile)

Max Activating Token Index: 710

Text #10

Max Range: 11.0219. Min Range: -11.0219

Max Act: 6.7564. Min Act: -0.0000

Data Index: 257639 (The Pile)

Max Activating Token Index: 937

Text #11

Max Range: 11.0219. Min Range: -11.0219

Max Act: 6.4917. Min Act: 0.0000

Data Index: 582580 (The Pile)

Max Activating Token Index: 694

Text #12

Max Range: 11.0219. Min Range: -11.0219

Max Act: 6.5152. Min Act: 0.0000

Data Index: 394055 (The Pile)

Max Activating Token Index: 84

Text #13

Max Range: 11.0219. Min Range: -11.0219

Max Act: 5.7240. Min Act: -0.0000

Data Index: 4253 (The Pile)

Max Activating Token Index: 246

Text #14

Max Range: 11.0219. Min Range: -11.0219

Max Act: 6.0555. Min Act: -0.0000

Data Index: 1854948 (The Pile)

Max Activating Token Index: 195

Text #15

Max Range: 11.0219. Min Range: -11.0219

Max Act: 6.2156. Min Act: 0.0000

Data Index: 572620 (The Pile)

Max Activating Token Index: 256

Text #16

Max Range: 11.0219. Min Range: -11.0219

Max Act: 6.7762. Min Act: -0.0000

Data Index: 217834 (The Pile)

Max Activating Token Index: 162

Text #17

Max Range: 11.0219. Min Range: -11.0219

Max Act: 5.9237. Min Act: 0.0000

Data Index: 1989895 (The Pile)

Max Activating Token Index: 107

Text #18

Max Range: 11.0219. Min Range: -11.0219

Max Act: 5.1742. Min Act: -0.0000

Data Index: 1991820 (The Pile)

Max Activating Token Index: 734

Text #19

Max Range: 11.0219. Min Range: -11.0219

Max Act: 5.5833. Min Act: 0.0000

Data Index: 995321 (The Pile)

Max Activating Token Index: 353