Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2281 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 4.5908. Min Range: -4.5908

Max Act: 4.5908. Min Act: -0.0001

Data Index: 737252 (C4 (Web Text))

Max Activating Token Index: 929

Text #1

Max Range: 4.5908. Min Range: -4.5908

Max Act: 4.5007. Min Act: -0.0001

Data Index: 605868 (C4 (Web Text))

Max Activating Token Index: 969

Text #2

Max Range: 4.5908. Min Range: -4.5908

Max Act: 4.2972. Min Act: -0.0001

Data Index: 946520 (C4 (Web Text))

Max Activating Token Index: 340

Text #3

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.8313. Min Act: -0.0001

Data Index: 695685 (C4 (Web Text))

Max Activating Token Index: 927

Text #4

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.4204. Min Act: -0.0001

Data Index: 1313183 (C4 (Web Text))

Max Activating Token Index: 741

Text #5

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.9936. Min Act: -0.0001

Data Index: 399288 (C4 (Web Text))

Max Activating Token Index: 39

Text #6

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.6138. Min Act: -0.0001

Data Index: 1275718 (C4 (Web Text))

Max Activating Token Index: 1005

Text #7

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.7219. Min Act: -0.0001

Data Index: 1137440 (C4 (Web Text))

Max Activating Token Index: 979

Text #8

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.6535. Min Act: -0.0001

Data Index: 1046555 (C4 (Web Text))

Max Activating Token Index: 350

Text #9

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.4629. Min Act: -0.0001

Data Index: 789334 (C4 (Web Text))

Max Activating Token Index: 483

Text #10

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.1898. Min Act: -0.0001

Data Index: 293645 (C4 (Web Text))

Max Activating Token Index: 868

Text #11

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.3928. Min Act: -0.0001

Data Index: 160870 (C4 (Web Text))

Max Activating Token Index: 553

Text #12

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.3187. Min Act: -0.0001

Data Index: 904476 (C4 (Web Text))

Max Activating Token Index: 291

Text #13

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.5311. Min Act: -0.0001

Data Index: 940000 (C4 (Web Text))

Max Activating Token Index: 826

Text #14

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.4822. Min Act: -0.0001

Data Index: 101248 (C4 (Web Text))

Max Activating Token Index: 47

Text #15

Max Range: 4.5908. Min Range: -4.5908

Max Act: 2.9494. Min Act: -0.0001

Data Index: 693066 (C4 (Web Text))

Max Activating Token Index: 218

Text #16

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.6758. Min Act: -0.0001

Data Index: 926652 (C4 (Web Text))

Max Activating Token Index: 364

Text #17

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.6178. Min Act: -0.0001

Data Index: 784410 (C4 (Web Text))

Max Activating Token Index: 621

Text #18

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.2026. Min Act: -0.0001

Data Index: 1183731 (C4 (Web Text))

Max Activating Token Index: 495

Text #19

Max Range: 4.5908. Min Range: -4.5908

Max Act: 3.2971. Min Act: -0.0001

Data Index: 1132074 (C4 (Web Text))

Max Activating Token Index: 515