Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2005 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2631. Min Act: -0.0001

Data Index: 1336 (C4 (Web Text))

Max Activating Token Index: 595

Text #1

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2768. Min Act: -0.0001

Data Index: 129634 (C4 (Web Text))

Max Activating Token Index: 942

Text #2

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2659. Min Act: -0.0001

Data Index: 540355 (C4 (Web Text))

Max Activating Token Index: 771

Text #3

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2628. Min Act: -0.0001

Data Index: 71668 (C4 (Web Text))

Max Activating Token Index: 592

Text #4

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2537. Min Act: -0.0001

Data Index: 294363 (C4 (Web Text))

Max Activating Token Index: 143

Text #5

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2612. Min Act: -0.0001

Data Index: 1329286 (C4 (Web Text))

Max Activating Token Index: 711

Text #6

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2533. Min Act: -0.0000

Data Index: 596792 (C4 (Web Text))

Max Activating Token Index: 875

Text #7

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2589. Min Act: -0.0001

Data Index: 484323 (C4 (Web Text))

Max Activating Token Index: 567

Text #8

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2428. Min Act: -0.0000

Data Index: 753037 (C4 (Web Text))

Max Activating Token Index: 239

Text #9

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2423. Min Act: -0.0001

Data Index: 136096 (C4 (Web Text))

Max Activating Token Index: 741

Text #10

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2395. Min Act: -0.0001

Data Index: 855076 (C4 (Web Text))

Max Activating Token Index: 757

Text #11

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2455. Min Act: -0.0001

Data Index: 733762 (C4 (Web Text))

Max Activating Token Index: 522

Text #12

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2528. Min Act: -0.0001

Data Index: 184113 (C4 (Web Text))

Max Activating Token Index: 441

Text #13

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2543. Min Act: -0.0001

Data Index: 343472 (C4 (Web Text))

Max Activating Token Index: 782

Text #14

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2442. Min Act: -0.0001

Data Index: 482999 (C4 (Web Text))

Max Activating Token Index: 841

Text #15

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2451. Min Act: -0.0001

Data Index: 916442 (C4 (Web Text))

Max Activating Token Index: 794

Text #16

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2523. Min Act: -0.0001

Data Index: 1173239 (C4 (Web Text))

Max Activating Token Index: 963

Text #17

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2443. Min Act: -0.0001

Data Index: 39789 (C4 (Web Text))

Max Activating Token Index: 660

Text #18

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2365. Min Act: -0.0001

Data Index: 841384 (C4 (Web Text))

Max Activating Token Index: 571

Text #19

Max Range: 0.2768. Min Range: -0.2768

Max Act: 0.2413. Min Act: -0.0001

Data Index: 29161 (C4 (Web Text))

Max Activating Token Index: 1006