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

Dataset: The Pile

Neuron 5096 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.9887. Min Act: -0.0000

Data Index: 1753999 (The Pile)

Max Activating Token Index: 222

Text #1

Max Range: 5.2168. Min Range: -5.2168

Max Act: 5.2168. Min Act: -0.0000

Data Index: 1119659 (The Pile)

Max Activating Token Index: 206

Text #2

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.9720. Min Act: -0.0000

Data Index: 1406621 (The Pile)

Max Activating Token Index: 755

Text #3

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.9391. Min Act: -0.0000

Data Index: 1013424 (The Pile)

Max Activating Token Index: 655

Text #4

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.5910. Min Act: -0.0000

Data Index: 571987 (The Pile)

Max Activating Token Index: 123

Text #5

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.6564. Min Act: -0.0000

Data Index: 615652 (The Pile)

Max Activating Token Index: 140

Text #6

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.8826. Min Act: -0.0000

Data Index: 511620 (The Pile)

Max Activating Token Index: 550

Text #7

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.7936. Min Act: -0.0000

Data Index: 1452726 (The Pile)

Max Activating Token Index: 182

Text #8

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.8207. Min Act: -0.0000

Data Index: 1851474 (The Pile)

Max Activating Token Index: 936

Text #9

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.5842. Min Act: -0.0000

Data Index: 832836 (The Pile)

Max Activating Token Index: 136

Text #10

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.9045. Min Act: -0.0000

Data Index: 193726 (The Pile)

Max Activating Token Index: 936

Text #11

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.2856. Min Act: -0.0000

Data Index: 84324 (The Pile)

Max Activating Token Index: 724

Text #12

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.4845. Min Act: -0.0000

Data Index: 1726064 (The Pile)

Max Activating Token Index: 441

Text #13

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.4829. Min Act: -0.0000

Data Index: 674676 (The Pile)

Max Activating Token Index: 74

Text #14

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.7983. Min Act: -0.0000

Data Index: 866918 (The Pile)

Max Activating Token Index: 169

Text #15

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.5653. Min Act: -0.0000

Data Index: 526828 (The Pile)

Max Activating Token Index: 472

Text #16

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.5989. Min Act: -0.0000

Data Index: 462276 (The Pile)

Max Activating Token Index: 814

Text #17

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.3080. Min Act: -0.0000

Data Index: 513936 (The Pile)

Max Activating Token Index: 221

Text #18

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.3459. Min Act: -0.0000

Data Index: 862782 (The Pile)

Max Activating Token Index: 54

Text #19

Max Range: 5.2168. Min Range: -5.2168

Max Act: 4.3000. Min Act: -0.0000

Data Index: 306489 (The Pile)

Max Activating Token Index: 227