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

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

Neuron 3927 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.5508. Min Act: -0.0000

Data Index: 252072 (C4 (Web Text))

Max Activating Token Index: 76

Text #1

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.3438. Min Act: -0.0000

Data Index: 407058 (C4 (Web Text))

Max Activating Token Index: 1021

Text #2

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.3640. Min Act: -0.0000

Data Index: 453754 (C4 (Web Text))

Max Activating Token Index: 1000

Text #3

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.3293. Min Act: -0.0000

Data Index: 589935 (C4 (Web Text))

Max Activating Token Index: 958

Text #4

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.3033. Min Act: -0.0000

Data Index: 1228461 (C4 (Web Text))

Max Activating Token Index: 175

Text #5

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.3141. Min Act: -0.0000

Data Index: 1174140 (C4 (Web Text))

Max Activating Token Index: 721

Text #6

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2826. Min Act: -0.0000

Data Index: 828371 (C4 (Web Text))

Max Activating Token Index: 657

Text #7

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.3003. Min Act: -0.0000

Data Index: 589510 (C4 (Web Text))

Max Activating Token Index: 1011

Text #8

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.3218. Min Act: -0.0000

Data Index: 1606693 (Python Code)

Max Activating Token Index: 869

Text #9

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2897. Min Act: -0.0000

Data Index: 634356 (C4 (Web Text))

Max Activating Token Index: 623

Text #10

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2892. Min Act: -0.0000

Data Index: 1189530 (C4 (Web Text))

Max Activating Token Index: 951

Text #11

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2864. Min Act: -0.0000

Data Index: 395109 (C4 (Web Text))

Max Activating Token Index: 439

Text #12

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2677. Min Act: -0.0000

Data Index: 1530466 (Python Code)

Max Activating Token Index: 255

Text #13

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2789. Min Act: -0.0000

Data Index: 52462 (C4 (Web Text))

Max Activating Token Index: 601

Text #14

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2882. Min Act: -0.0000

Data Index: 739394 (C4 (Web Text))

Max Activating Token Index: 429

Text #15

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2683. Min Act: -0.0000

Data Index: 1114916 (C4 (Web Text))

Max Activating Token Index: 758

Text #16

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2739. Min Act: -0.0000

Data Index: 632185 (C4 (Web Text))

Max Activating Token Index: 390

Text #17

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2603. Min Act: -0.0000

Data Index: 1357895 (C4 (Web Text))

Max Activating Token Index: 74

Text #18

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2512. Min Act: -0.0000

Data Index: 1105761 (C4 (Web Text))

Max Activating Token Index: 879

Text #19

Max Range: 0.5508. Min Range: -0.5508

Max Act: 0.2484. Min Act: -0.0000

Data Index: 249388 (C4 (Web Text))

Max Activating Token Index: 125