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

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 4969 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.0499. Min Range: -5.0499

Max Act: 5.0499. Min Act: -0.0000

Data Index: 380740 (C4 (Web Text))

Max Activating Token Index: 546

Text #1

Max Range: 5.0499. Min Range: -5.0499

Max Act: 4.5247. Min Act: -0.0000

Data Index: 777698 (C4 (Web Text))

Max Activating Token Index: 965

Text #2

Max Range: 5.0499. Min Range: -5.0499

Max Act: 3.6406. Min Act: -0.0000

Data Index: 98800 (C4 (Web Text))

Max Activating Token Index: 1017

Text #3

Max Range: 5.0499. Min Range: -5.0499

Max Act: 3.3414. Min Act: -0.0000

Data Index: 1149848 (C4 (Web Text))

Max Activating Token Index: 673

Text #4

Max Range: 5.0499. Min Range: -5.0499

Max Act: 3.1075. Min Act: -0.0000

Data Index: 671360 (C4 (Web Text))

Max Activating Token Index: 947

Text #5

Max Range: 5.0499. Min Range: -5.0499

Max Act: 3.4073. Min Act: -0.0000

Data Index: 427864 (C4 (Web Text))

Max Activating Token Index: 237

Text #6

Max Range: 5.0499. Min Range: -5.0499

Max Act: 3.3391. Min Act: -0.0000

Data Index: 406699 (C4 (Web Text))

Max Activating Token Index: 549

Text #7

Max Range: 5.0499. Min Range: -5.0499

Max Act: 3.0355. Min Act: -0.0000

Data Index: 1337966 (C4 (Web Text))

Max Activating Token Index: 442

Text #8

Max Range: 5.0499. Min Range: -5.0499

Max Act: 3.0122. Min Act: -0.0000

Data Index: 380105 (C4 (Web Text))

Max Activating Token Index: 959

Text #9

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.8479. Min Act: -0.0000

Data Index: 160466 (C4 (Web Text))

Max Activating Token Index: 120

Text #10

Max Range: 5.0499. Min Range: -5.0499

Max Act: 3.0097. Min Act: -0.0000

Data Index: 190665 (C4 (Web Text))

Max Activating Token Index: 807

Text #11

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.6929. Min Act: -0.0000

Data Index: 336394 (C4 (Web Text))

Max Activating Token Index: 594

Text #12

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.7866. Min Act: -0.0000

Data Index: 1380479 (Python Code)

Max Activating Token Index: 774

Text #13

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.5297. Min Act: -0.0000

Data Index: 364060 (C4 (Web Text))

Max Activating Token Index: 728

Text #14

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.4039. Min Act: -0.0000

Data Index: 1213552 (C4 (Web Text))

Max Activating Token Index: 762

Text #15

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.6210. Min Act: -0.0000

Data Index: 317254 (C4 (Web Text))

Max Activating Token Index: 375

Text #16

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.5998. Min Act: -0.0000

Data Index: 1202366 (C4 (Web Text))

Max Activating Token Index: 700

Text #17

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.5609. Min Act: -0.0000

Data Index: 652139 (C4 (Web Text))

Max Activating Token Index: 212

Text #18

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.3789. Min Act: -0.0000

Data Index: 870658 (C4 (Web Text))

Max Activating Token Index: 345

Text #19

Max Range: 5.0499. Min Range: -5.0499

Max Act: 2.3793. Min Act: -0.0000

Data Index: 1136893 (C4 (Web Text))

Max Activating Token Index: 918