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

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

Neuron 22 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.9396. Min Act: -0.0001

Data Index: 1352162 (C4 (Web Text))

Max Activating Token Index: 826

Text #1

Max Range: 1.0064. Min Range: -1.0064

Max Act: 1.0064. Min Act: -0.0001

Data Index: 708943 (C4 (Web Text))

Max Activating Token Index: 32

Text #2

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.9907. Min Act: -0.0001

Data Index: 162054 (C4 (Web Text))

Max Activating Token Index: 949

Text #3

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.9629. Min Act: -0.0001

Data Index: 484702 (C4 (Web Text))

Max Activating Token Index: 26

Text #4

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.9595. Min Act: -0.0001

Data Index: 605444 (C4 (Web Text))

Max Activating Token Index: 119

Text #5

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.9037. Min Act: -0.0001

Data Index: 6066 (C4 (Web Text))

Max Activating Token Index: 552

Text #6

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.8178. Min Act: -0.0001

Data Index: 1168721 (C4 (Web Text))

Max Activating Token Index: 116

Text #7

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7939. Min Act: -0.0001

Data Index: 1262515 (C4 (Web Text))

Max Activating Token Index: 494

Text #8

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7735. Min Act: -0.0001

Data Index: 1223858 (C4 (Web Text))

Max Activating Token Index: 416

Text #9

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7983. Min Act: -0.0001

Data Index: 295110 (C4 (Web Text))

Max Activating Token Index: 17

Text #10

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7956. Min Act: -0.0001

Data Index: 314484 (C4 (Web Text))

Max Activating Token Index: 679

Text #11

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7348. Min Act: -0.0001

Data Index: 1163975 (C4 (Web Text))

Max Activating Token Index: 1001

Text #12

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7893. Min Act: -0.0001

Data Index: 1139629 (C4 (Web Text))

Max Activating Token Index: 667

Text #13

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7273. Min Act: -0.0001

Data Index: 807044 (C4 (Web Text))

Max Activating Token Index: 951

Text #14

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7982. Min Act: -0.0001

Data Index: 1055025 (C4 (Web Text))

Max Activating Token Index: 87

Text #15

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7707. Min Act: -0.0001

Data Index: 59473 (C4 (Web Text))

Max Activating Token Index: 923

Text #16

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7611. Min Act: -0.0001

Data Index: 10389 (C4 (Web Text))

Max Activating Token Index: 1018

Text #17

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7591. Min Act: -0.0001

Data Index: 220097 (C4 (Web Text))

Max Activating Token Index: 718

Text #18

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7569. Min Act: -0.0001

Data Index: 410320 (C4 (Web Text))

Max Activating Token Index: 296

Text #19

Max Range: 1.0064. Min Range: -1.0064

Max Act: 0.7765. Min Act: -0.0001

Data Index: 945785 (C4 (Web Text))

Max Activating Token Index: 4