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

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

Neuron 4223 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3046. Min Range: -3.3046

Max Act: 3.3046. Min Act: -0.0000

Data Index: 70765 (C4 (Web Text))

Max Activating Token Index: 977

Text #1

Max Range: 3.3046. Min Range: -3.3046

Max Act: 3.1708. Min Act: -0.0000

Data Index: 946464 (C4 (Web Text))

Max Activating Token Index: 661

Text #2

Max Range: 3.3046. Min Range: -3.3046

Max Act: 3.1164. Min Act: -0.0000

Data Index: 1513762 (Python Code)

Max Activating Token Index: 10

Text #3

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.8109. Min Act: -0.0000

Data Index: 679961 (C4 (Web Text))

Max Activating Token Index: 285

Text #4

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.8374. Min Act: -0.0000

Data Index: 479733 (C4 (Web Text))

Max Activating Token Index: 465

Text #5

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.9465. Min Act: -0.0000

Data Index: 1353635 (C4 (Web Text))

Max Activating Token Index: 46

Text #6

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.9577. Min Act: -0.0000

Data Index: 659684 (C4 (Web Text))

Max Activating Token Index: 36

Text #7

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.4686. Min Act: -0.0000

Data Index: 922172 (C4 (Web Text))

Max Activating Token Index: 868

Text #8

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.5652. Min Act: -0.0000

Data Index: 226856 (C4 (Web Text))

Max Activating Token Index: 157

Text #9

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.5877. Min Act: -0.0000

Data Index: 400862 (C4 (Web Text))

Max Activating Token Index: 767

Text #10

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.3935. Min Act: -0.0000

Data Index: 48735 (C4 (Web Text))

Max Activating Token Index: 633

Text #11

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.4881. Min Act: -0.0000

Data Index: 353948 (C4 (Web Text))

Max Activating Token Index: 780

Text #12

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.4647. Min Act: -0.0000

Data Index: 1210654 (C4 (Web Text))

Max Activating Token Index: 846

Text #13

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.4541. Min Act: -0.0000

Data Index: 1431557 (Python Code)

Max Activating Token Index: 34

Text #14

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.4374. Min Act: -0.0000

Data Index: 203424 (C4 (Web Text))

Max Activating Token Index: 62

Text #15

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.4273. Min Act: -0.0000

Data Index: 495693 (C4 (Web Text))

Max Activating Token Index: 928

Text #16

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.4369. Min Act: -0.0000

Data Index: 1269154 (C4 (Web Text))

Max Activating Token Index: 905

Text #17

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.4375. Min Act: -0.0000

Data Index: 837302 (C4 (Web Text))

Max Activating Token Index: 875

Text #18

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.2656. Min Act: -0.0000

Data Index: 127311 (C4 (Web Text))

Max Activating Token Index: 315

Text #19

Max Range: 3.3046. Min Range: -3.3046

Max Act: 2.3497. Min Act: -0.0000

Data Index: 1152396 (C4 (Web Text))

Max Activating Token Index: 248