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

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

Neuron 3077 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.3596. Min Act: -0.0000

Data Index: 1549071 (Python Code)

Max Activating Token Index: 796

Text #1

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.3174. Min Act: -0.0000

Data Index: 1258517 (C4 (Web Text))

Max Activating Token Index: 217

Text #2

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2807. Min Act: -0.0000

Data Index: 712819 (C4 (Web Text))

Max Activating Token Index: 563

Text #3

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2818. Min Act: -0.0000

Data Index: 340335 (C4 (Web Text))

Max Activating Token Index: 361

Text #4

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2848. Min Act: -0.0000

Data Index: 1576596 (Python Code)

Max Activating Token Index: 130

Text #5

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2712. Min Act: -0.0000

Data Index: 747530 (C4 (Web Text))

Max Activating Token Index: 155

Text #6

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2679. Min Act: -0.0000

Data Index: 647189 (C4 (Web Text))

Max Activating Token Index: 926

Text #7

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2650. Min Act: -0.0000

Data Index: 155852 (C4 (Web Text))

Max Activating Token Index: 945

Text #8

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2531. Min Act: -0.0000

Data Index: 763929 (C4 (Web Text))

Max Activating Token Index: 944

Text #9

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2589. Min Act: -0.0000

Data Index: 1537062 (Python Code)

Max Activating Token Index: 56

Text #10

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2439. Min Act: -0.0000

Data Index: 792533 (C4 (Web Text))

Max Activating Token Index: 707

Text #11

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2507. Min Act: -0.0000

Data Index: 832405 (C4 (Web Text))

Max Activating Token Index: 1000

Text #12

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2501. Min Act: -0.0000

Data Index: 844348 (C4 (Web Text))

Max Activating Token Index: 355

Text #13

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2348. Min Act: -0.0000

Data Index: 1373496 (Python Code)

Max Activating Token Index: 704

Text #14

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2321. Min Act: -0.0000

Data Index: 1215465 (C4 (Web Text))

Max Activating Token Index: 672

Text #15

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2466. Min Act: -0.0000

Data Index: 69019 (C4 (Web Text))

Max Activating Token Index: 964

Text #16

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2398. Min Act: -0.0000

Data Index: 527346 (C4 (Web Text))

Max Activating Token Index: 988

Text #17

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2345. Min Act: -0.0000

Data Index: 893063 (C4 (Web Text))

Max Activating Token Index: 832

Text #18

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2210. Min Act: -0.0000

Data Index: 1560579 (Python Code)

Max Activating Token Index: 912

Text #19

Max Range: 0.3596. Min Range: -0.3596

Max Act: 0.2112. Min Act: -0.0000

Data Index: 710716 (C4 (Web Text))

Max Activating Token Index: 58