Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4202 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8912. Min Act: -0.0001

Data Index: 214241 (C4 (Web Text))

Max Activating Token Index: 24

Text #1

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8477. Min Act: -0.0001

Data Index: 957302 (C4 (Web Text))

Max Activating Token Index: 993

Text #2

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8597. Min Act: -0.0001

Data Index: 1115350 (C4 (Web Text))

Max Activating Token Index: 679

Text #3

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8325. Min Act: -0.0001

Data Index: 271321 (C4 (Web Text))

Max Activating Token Index: 288

Text #4

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8138. Min Act: -0.0001

Data Index: 850177 (C4 (Web Text))

Max Activating Token Index: 96

Text #5

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8509. Min Act: -0.0001

Data Index: 1136728 (C4 (Web Text))

Max Activating Token Index: 18

Text #6

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.7938. Min Act: -0.0001

Data Index: 1193896 (C4 (Web Text))

Max Activating Token Index: 921

Text #7

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.7773. Min Act: -0.0001

Data Index: 266205 (C4 (Web Text))

Max Activating Token Index: 87

Text #8

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8117. Min Act: -0.0001

Data Index: 1044049 (C4 (Web Text))

Max Activating Token Index: 65

Text #9

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8041. Min Act: -0.0001

Data Index: 728686 (C4 (Web Text))

Max Activating Token Index: 538

Text #10

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.7724. Min Act: -0.0001

Data Index: 719733 (C4 (Web Text))

Max Activating Token Index: 888

Text #11

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8256. Min Act: -0.0001

Data Index: 704833 (C4 (Web Text))

Max Activating Token Index: 93

Text #12

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.7145. Min Act: -0.0001

Data Index: 746493 (C4 (Web Text))

Max Activating Token Index: 743

Text #13

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.6939. Min Act: -0.0001

Data Index: 1336418 (C4 (Web Text))

Max Activating Token Index: 253

Text #14

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.7718. Min Act: -0.0001

Data Index: 720358 (C4 (Web Text))

Max Activating Token Index: 104

Text #15

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.7263. Min Act: -0.0001

Data Index: 1628578 (Python Code)

Max Activating Token Index: 882

Text #16

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.7205. Min Act: -0.0001

Data Index: 44557 (C4 (Web Text))

Max Activating Token Index: 12

Text #17

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.7111. Min Act: -0.0001

Data Index: 1651467 (Python Code)

Max Activating Token Index: 886

Text #18

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.7814. Min Act: -0.0001

Data Index: 15662 (C4 (Web Text))

Max Activating Token Index: 751

Text #19

Max Range: 0.8912. Min Range: -0.8912

Max Act: 0.8098. Min Act: -0.0001

Data Index: 1293587 (C4 (Web Text))

Max Activating Token Index: 809