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

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

Neuron 1035 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.5525. Min Act: -0.0000

Data Index: 52200 (C4 (Web Text))

Max Activating Token Index: 94

Text #1

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.4232. Min Act: -0.0000

Data Index: 406994 (C4 (Web Text))

Max Activating Token Index: 64

Text #2

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.4018. Min Act: -0.0000

Data Index: 506018 (C4 (Web Text))

Max Activating Token Index: 915

Text #3

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.3295. Min Act: -0.0000

Data Index: 1357457 (C4 (Web Text))

Max Activating Token Index: 121

Text #4

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.3269. Min Act: -0.0001

Data Index: 1154834 (C4 (Web Text))

Max Activating Token Index: 156

Text #5

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.3349. Min Act: -0.0000

Data Index: 1351121 (C4 (Web Text))

Max Activating Token Index: 105

Text #6

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2625. Min Act: -0.0001

Data Index: 1043944 (C4 (Web Text))

Max Activating Token Index: 778

Text #7

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2880. Min Act: -0.0000

Data Index: 1421797 (Python Code)

Max Activating Token Index: 753

Text #8

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2599. Min Act: -0.0000

Data Index: 685183 (C4 (Web Text))

Max Activating Token Index: 688

Text #9

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2821. Min Act: -0.0000

Data Index: 948170 (C4 (Web Text))

Max Activating Token Index: 689

Text #10

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2377. Min Act: -0.0001

Data Index: 1377613 (Python Code)

Max Activating Token Index: 817

Text #11

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2641. Min Act: -0.0000

Data Index: 594218 (C4 (Web Text))

Max Activating Token Index: 530

Text #12

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2515. Min Act: -0.0001

Data Index: 1378451 (Python Code)

Max Activating Token Index: 709

Text #13

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2557. Min Act: -0.0000

Data Index: 276176 (C4 (Web Text))

Max Activating Token Index: 1005

Text #14

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2529. Min Act: -0.0001

Data Index: 1447369 (Python Code)

Max Activating Token Index: 353

Text #15

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2886. Min Act: -0.0000

Data Index: 968145 (C4 (Web Text))

Max Activating Token Index: 386

Text #16

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2479. Min Act: -0.0000

Data Index: 1593393 (Python Code)

Max Activating Token Index: 478

Text #17

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2353. Min Act: -0.0000

Data Index: 1505276 (Python Code)

Max Activating Token Index: 766

Text #18

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2553. Min Act: -0.0000

Data Index: 1387059 (Python Code)

Max Activating Token Index: 378

Text #19

Max Range: 0.5525. Min Range: -0.5525

Max Act: 0.2183. Min Act: -0.0000

Data Index: 916731 (C4 (Web Text))

Max Activating Token Index: 608