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

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

Neuron 614 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6795. Min Act: -0.0003

Data Index: 608859 (C4 (Web Text))

Max Activating Token Index: 157

Text #1

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6924. Min Act: -0.0003

Data Index: 852772 (C4 (Web Text))

Max Activating Token Index: 183

Text #2

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6518. Min Act: -0.0003

Data Index: 314585 (C4 (Web Text))

Max Activating Token Index: 255

Text #3

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6527. Min Act: -0.0003

Data Index: 591131 (C4 (Web Text))

Max Activating Token Index: 60

Text #4

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6296. Min Act: -0.0003

Data Index: 550414 (C4 (Web Text))

Max Activating Token Index: 701

Text #5

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.5960. Min Act: -0.0003

Data Index: 119027 (C4 (Web Text))

Max Activating Token Index: 953

Text #6

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6177. Min Act: -0.0003

Data Index: 938221 (C4 (Web Text))

Max Activating Token Index: 989

Text #7

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6379. Min Act: -0.0003

Data Index: 807412 (C4 (Web Text))

Max Activating Token Index: 63

Text #8

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.5923. Min Act: -0.0003

Data Index: 346066 (C4 (Web Text))

Max Activating Token Index: 899

Text #9

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6114. Min Act: -0.0003

Data Index: 199982 (C4 (Web Text))

Max Activating Token Index: 32

Text #10

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.5662. Min Act: -0.0003

Data Index: 1237601 (C4 (Web Text))

Max Activating Token Index: 258

Text #11

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.5943. Min Act: -0.0003

Data Index: 360547 (C4 (Web Text))

Max Activating Token Index: 991

Text #12

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.5888. Min Act: -0.0003

Data Index: 229801 (C4 (Web Text))

Max Activating Token Index: 775

Text #13

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.5623. Min Act: -0.0003

Data Index: 56244 (C4 (Web Text))

Max Activating Token Index: 67

Text #14

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.5629. Min Act: -0.0003

Data Index: 988915 (C4 (Web Text))

Max Activating Token Index: 920

Text #15

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6070. Min Act: -0.0003

Data Index: 695658 (C4 (Web Text))

Max Activating Token Index: 900

Text #16

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.5216. Min Act: -0.0003

Data Index: 36232 (C4 (Web Text))

Max Activating Token Index: 319

Text #17

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.5588. Min Act: -0.0003

Data Index: 1128348 (C4 (Web Text))

Max Activating Token Index: 690

Text #18

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6144. Min Act: -0.0003

Data Index: 465854 (C4 (Web Text))

Max Activating Token Index: 509

Text #19

Max Range: 0.6924. Min Range: -0.6924

Max Act: 0.6047. Min Act: -0.0003

Data Index: 65000 (C4 (Web Text))

Max Activating Token Index: 797