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

Dataset: The Pile

Neuron 3334 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.7746. Min Act: -0.0000

Data Index: 633407 (The Pile)

Max Activating Token Index: 129

Text #1

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.8400. Min Act: -0.0000

Data Index: 1609857 (The Pile)

Max Activating Token Index: 627

Text #2

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.8260. Min Act: -0.0000

Data Index: 1012918 (The Pile)

Max Activating Token Index: 786

Text #3

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.7968. Min Act: -0.0000

Data Index: 1384161 (The Pile)

Max Activating Token Index: 985

Text #4

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.6872. Min Act: -0.0000

Data Index: 138358 (The Pile)

Max Activating Token Index: 218

Text #5

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.5542. Min Act: -0.0000

Data Index: 861345 (The Pile)

Max Activating Token Index: 349

Text #6

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.6302. Min Act: -0.0000

Data Index: 331674 (The Pile)

Max Activating Token Index: 108

Text #7

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.6069. Min Act: -0.0000

Data Index: 654040 (The Pile)

Max Activating Token Index: 875

Text #8

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.4281. Min Act: -0.0000

Data Index: 271924 (The Pile)

Max Activating Token Index: 770

Text #9

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.6809. Min Act: -0.0000

Data Index: 717543 (The Pile)

Max Activating Token Index: 343

Text #10

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.5702. Min Act: -0.0000

Data Index: 1756 (The Pile)

Max Activating Token Index: 190

Text #11

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.4739. Min Act: -0.0000

Data Index: 688829 (The Pile)

Max Activating Token Index: 287

Text #12

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.4961. Min Act: -0.0000

Data Index: 102036 (The Pile)

Max Activating Token Index: 748

Text #13

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.4889. Min Act: -0.0000

Data Index: 59075 (The Pile)

Max Activating Token Index: 210

Text #14

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.4052. Min Act: -0.0000

Data Index: 1930018 (The Pile)

Max Activating Token Index: 328

Text #15

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.4788. Min Act: -0.0000

Data Index: 1568629 (The Pile)

Max Activating Token Index: 54

Text #16

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.4745. Min Act: -0.0000

Data Index: 1177506 (The Pile)

Max Activating Token Index: 154

Text #17

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.4782. Min Act: -0.0000

Data Index: 803881 (The Pile)

Max Activating Token Index: 958

Text #18

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.3796. Min Act: -0.0000

Data Index: 892863 (The Pile)

Max Activating Token Index: 537

Text #19

Max Range: 1.8400. Min Range: -1.8400

Max Act: 1.4652. Min Act: -0.0000

Data Index: 1578958 (The Pile)

Max Activating Token Index: 194