Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2673 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.3818. Min Range: -2.3818

Max Act: 2.3818. Min Act: -0.0000

Data Index: 427904 (C4 (Web Text))

Max Activating Token Index: 5

Text #1

Max Range: 2.3818. Min Range: -2.3818

Max Act: 2.0843. Min Act: -0.0000

Data Index: 1447037 (Python Code)

Max Activating Token Index: 90

Text #2

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.9807. Min Act: -0.0000

Data Index: 739704 (C4 (Web Text))

Max Activating Token Index: 19

Text #3

Max Range: 2.3818. Min Range: -2.3818

Max Act: 2.0916. Min Act: -0.0000

Data Index: 1590694 (Python Code)

Max Activating Token Index: 118

Text #4

Max Range: 2.3818. Min Range: -2.3818

Max Act: 2.0659. Min Act: -0.0000

Data Index: 1493802 (Python Code)

Max Activating Token Index: 515

Text #5

Max Range: 2.3818. Min Range: -2.3818

Max Act: 2.0463. Min Act: -0.0000

Data Index: 599952 (C4 (Web Text))

Max Activating Token Index: 115

Text #6

Max Range: 2.3818. Min Range: -2.3818

Max Act: 2.0350. Min Act: -0.0000

Data Index: 500056 (C4 (Web Text))

Max Activating Token Index: 239

Text #7

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.9680. Min Act: -0.0000

Data Index: 64989 (C4 (Web Text))

Max Activating Token Index: 79

Text #8

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.9520. Min Act: -0.0000

Data Index: 1478087 (Python Code)

Max Activating Token Index: 378

Text #9

Max Range: 2.3818. Min Range: -2.3818

Max Act: 2.0437. Min Act: -0.0000

Data Index: 835079 (C4 (Web Text))

Max Activating Token Index: 616

Text #10

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.9147. Min Act: -0.0000

Data Index: 9066 (C4 (Web Text))

Max Activating Token Index: 949

Text #11

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.8869. Min Act: -0.0000

Data Index: 910250 (C4 (Web Text))

Max Activating Token Index: 880

Text #12

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.8891. Min Act: -0.0000

Data Index: 133341 (C4 (Web Text))

Max Activating Token Index: 585

Text #13

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.8649. Min Act: -0.0000

Data Index: 239112 (C4 (Web Text))

Max Activating Token Index: 568

Text #14

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.7376. Min Act: -0.0000

Data Index: 1436054 (Python Code)

Max Activating Token Index: 753

Text #15

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.8232. Min Act: -0.0000

Data Index: 1004270 (C4 (Web Text))

Max Activating Token Index: 762

Text #16

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.7693. Min Act: -0.0000

Data Index: 1595348 (Python Code)

Max Activating Token Index: 454

Text #17

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.6438. Min Act: -0.0000

Data Index: 185788 (C4 (Web Text))

Max Activating Token Index: 192

Text #18

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.7184. Min Act: -0.0000

Data Index: 565119 (C4 (Web Text))

Max Activating Token Index: 796

Text #19

Max Range: 2.3818. Min Range: -2.3818

Max Act: 1.6341. Min Act: -0.0000

Data Index: 858856 (C4 (Web Text))

Max Activating Token Index: 1017