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

Dataset: The Pile

Neuron 3896 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: 7.0835. Min Range: -7.0835

Max Act: 6.9676. Min Act: -0.0000

Data Index: 1140518 (The Pile)

Max Activating Token Index: 637

Text #1

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.8961. Min Act: -0.0000

Data Index: 692598 (The Pile)

Max Activating Token Index: 121

Text #2

Max Range: 7.0835. Min Range: -7.0835

Max Act: 7.0835. Min Act: -0.0000

Data Index: 1975555 (The Pile)

Max Activating Token Index: 180

Text #3

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.4746. Min Act: -0.0000

Data Index: 789079 (The Pile)

Max Activating Token Index: 876

Text #4

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.2200. Min Act: -0.0000

Data Index: 837258 (The Pile)

Max Activating Token Index: 269

Text #5

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.5784. Min Act: -0.0000

Data Index: 193705 (The Pile)

Max Activating Token Index: 88

Text #6

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.4885. Min Act: -0.0000

Data Index: 354224 (The Pile)

Max Activating Token Index: 766

Text #7

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.4330. Min Act: -0.0000

Data Index: 1468237 (The Pile)

Max Activating Token Index: 140

Text #8

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.7889. Min Act: -0.0000

Data Index: 1057222 (The Pile)

Max Activating Token Index: 113

Text #9

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.5135. Min Act: -0.0000

Data Index: 1247030 (The Pile)

Max Activating Token Index: 113

Text #10

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.3117. Min Act: -0.0000

Data Index: 1971834 (The Pile)

Max Activating Token Index: 930

Text #11

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.2711. Min Act: -0.0000

Data Index: 1642092 (The Pile)

Max Activating Token Index: 442

Text #12

Max Range: 7.0835. Min Range: -7.0835

Max Act: 5.9030. Min Act: -0.0000

Data Index: 897133 (The Pile)

Max Activating Token Index: 730

Text #13

Max Range: 7.0835. Min Range: -7.0835

Max Act: 5.6480. Min Act: -0.0000

Data Index: 1278409 (The Pile)

Max Activating Token Index: 792

Text #14

Max Range: 7.0835. Min Range: -7.0835

Max Act: 5.8255. Min Act: -0.0000

Data Index: 213211 (The Pile)

Max Activating Token Index: 693

Text #15

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.4151. Min Act: -0.0000

Data Index: 1563026 (The Pile)

Max Activating Token Index: 159

Text #16

Max Range: 7.0835. Min Range: -7.0835

Max Act: 5.6490. Min Act: -0.0000

Data Index: 1854364 (The Pile)

Max Activating Token Index: 939

Text #17

Max Range: 7.0835. Min Range: -7.0835

Max Act: 5.9760. Min Act: -0.0000

Data Index: 173164 (The Pile)

Max Activating Token Index: 848

Text #18

Max Range: 7.0835. Min Range: -7.0835

Max Act: 6.5724. Min Act: -0.0000

Data Index: 1390945 (The Pile)

Max Activating Token Index: 891

Text #19

Max Range: 7.0835. Min Range: -7.0835

Max Act: 5.4071. Min Act: -0.0000

Data Index: 1570283 (The Pile)

Max Activating Token Index: 873