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

Dataset: The Pile

Neuron 4077 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 3.2706. Min Range: -3.2706

Max Act: 3.2706. Min Act: -0.0000

Data Index: 1911856 (The Pile)

Max Activating Token Index: 727

Text #1

Max Range: 3.2706. Min Range: -3.2706

Max Act: 3.0746. Min Act: -0.0000

Data Index: 110268 (The Pile)

Max Activating Token Index: 610

Text #2

Max Range: 3.2706. Min Range: -3.2706

Max Act: 3.0289. Min Act: -0.0000

Data Index: 1428427 (The Pile)

Max Activating Token Index: 605

Text #3

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.6345. Min Act: -0.0000

Data Index: 1303435 (The Pile)

Max Activating Token Index: 947

Text #4

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.7064. Min Act: -0.0000

Data Index: 923411 (The Pile)

Max Activating Token Index: 415

Text #5

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.3031. Min Act: -0.0000

Data Index: 1876152 (The Pile)

Max Activating Token Index: 452

Text #6

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.3876. Min Act: -0.0000

Data Index: 1662515 (The Pile)

Max Activating Token Index: 584

Text #7

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.5432. Min Act: -0.0000

Data Index: 347948 (The Pile)

Max Activating Token Index: 600

Text #8

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.4802. Min Act: -0.0000

Data Index: 135497 (The Pile)

Max Activating Token Index: 382

Text #9

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.3173. Min Act: -0.0000

Data Index: 1182750 (The Pile)

Max Activating Token Index: 105

Text #10

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.3072. Min Act: -0.0000

Data Index: 1208647 (The Pile)

Max Activating Token Index: 289

Text #11

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.4070. Min Act: -0.0000

Data Index: 196503 (The Pile)

Max Activating Token Index: 445

Text #12

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.1263. Min Act: -0.0000

Data Index: 1306986 (The Pile)

Max Activating Token Index: 859

Text #13

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.1102. Min Act: -0.0000

Data Index: 831674 (The Pile)

Max Activating Token Index: 158

Text #14

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.2158. Min Act: -0.0000

Data Index: 311571 (The Pile)

Max Activating Token Index: 246

Text #15

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.2033. Min Act: -0.0000

Data Index: 836430 (The Pile)

Max Activating Token Index: 573

Text #16

Max Range: 3.2706. Min Range: -3.2706

Max Act: 1.9796. Min Act: -0.0000

Data Index: 572446 (The Pile)

Max Activating Token Index: 671

Text #17

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.2013. Min Act: -0.0000

Data Index: 124721 (The Pile)

Max Activating Token Index: 87

Text #18

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.3144. Min Act: -0.0000

Data Index: 632011 (The Pile)

Max Activating Token Index: 458

Text #19

Max Range: 3.2706. Min Range: -3.2706

Max Act: 2.0693. Min Act: -0.0000

Data Index: 1153733 (The Pile)

Max Activating Token Index: 472