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

Dataset: The Pile

Neuron 2057 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: 9.2701. Min Range: -9.2701

Max Act: 9.2475. Min Act: -0.0000

Data Index: 1313767 (The Pile)

Max Activating Token Index: 116

Text #1

Max Range: 9.2701. Min Range: -9.2701

Max Act: 9.2701. Min Act: -0.0000

Data Index: 739900 (The Pile)

Max Activating Token Index: 163

Text #2

Max Range: 9.2701. Min Range: -9.2701

Max Act: 9.1643. Min Act: -0.0000

Data Index: 1911744 (The Pile)

Max Activating Token Index: 470

Text #3

Max Range: 9.2701. Min Range: -9.2701

Max Act: 9.0863. Min Act: -0.0000

Data Index: 1110572 (The Pile)

Max Activating Token Index: 721

Text #4

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.6746. Min Act: -0.0000

Data Index: 593615 (The Pile)

Max Activating Token Index: 722

Text #5

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.2996. Min Act: -0.0000

Data Index: 436376 (The Pile)

Max Activating Token Index: 605

Text #6

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.6725. Min Act: -0.0000

Data Index: 1657065 (The Pile)

Max Activating Token Index: 975

Text #7

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.1575. Min Act: -0.0000

Data Index: 547451 (The Pile)

Max Activating Token Index: 754

Text #8

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.5409. Min Act: -0.0000

Data Index: 705746 (The Pile)

Max Activating Token Index: 309

Text #9

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.2854. Min Act: -0.0000

Data Index: 891300 (The Pile)

Max Activating Token Index: 452

Text #10

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.7314. Min Act: -0.0000

Data Index: 435420 (The Pile)

Max Activating Token Index: 914

Text #11

Max Range: 9.2701. Min Range: -9.2701

Max Act: 7.9056. Min Act: -0.0000

Data Index: 1367176 (The Pile)

Max Activating Token Index: 173

Text #12

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.7752. Min Act: -0.0000

Data Index: 91241 (The Pile)

Max Activating Token Index: 575

Text #13

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.3900. Min Act: -0.0000

Data Index: 1164451 (The Pile)

Max Activating Token Index: 824

Text #14

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.3794. Min Act: -0.0000

Data Index: 920079 (The Pile)

Max Activating Token Index: 471

Text #15

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.4688. Min Act: -0.0000

Data Index: 76513 (The Pile)

Max Activating Token Index: 469

Text #16

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.6961. Min Act: -0.0000

Data Index: 1194201 (The Pile)

Max Activating Token Index: 44

Text #17

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.4131. Min Act: -0.0000

Data Index: 1494808 (The Pile)

Max Activating Token Index: 855

Text #18

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.9821. Min Act: -0.0000

Data Index: 12557 (The Pile)

Max Activating Token Index: 252

Text #19

Max Range: 9.2701. Min Range: -9.2701

Max Act: 8.5715. Min Act: -0.0000

Data Index: 1211671 (The Pile)

Max Activating Token Index: 882