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

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

Neuron 3625 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: 0.1096. Min Range: -0.1096

Max Act: 0.1096. Min Act: -0.0001

Data Index: 1342493 (C4 (Web Text))

Max Activating Token Index: 308

Text #1

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0945. Min Act: -0.0001

Data Index: 1553655 (Python Code)

Max Activating Token Index: 154

Text #2

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0795. Min Act: -0.0001

Data Index: 554667 (C4 (Web Text))

Max Activating Token Index: 53

Text #3

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0749. Min Act: -0.0001

Data Index: 844230 (C4 (Web Text))

Max Activating Token Index: 565

Text #4

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0752. Min Act: -0.0001

Data Index: 1429074 (Python Code)

Max Activating Token Index: 734

Text #5

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0702. Min Act: -0.0001

Data Index: 564582 (C4 (Web Text))

Max Activating Token Index: 916

Text #6

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0723. Min Act: -0.0001

Data Index: 1141058 (C4 (Web Text))

Max Activating Token Index: 119

Text #7

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0715. Min Act: -0.0001

Data Index: 1499682 (Python Code)

Max Activating Token Index: 691

Text #8

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0665. Min Act: -0.0001

Data Index: 1019150 (C4 (Web Text))

Max Activating Token Index: 297

Text #9

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0683. Min Act: -0.0001

Data Index: 565624 (C4 (Web Text))

Max Activating Token Index: 346

Text #10

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0661. Min Act: -0.0001

Data Index: 86145 (C4 (Web Text))

Max Activating Token Index: 153

Text #11

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0663. Min Act: -0.0001

Data Index: 318300 (C4 (Web Text))

Max Activating Token Index: 33

Text #12

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0634. Min Act: -0.0001

Data Index: 885695 (C4 (Web Text))

Max Activating Token Index: 206

Text #13

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0668. Min Act: -0.0001

Data Index: 300798 (C4 (Web Text))

Max Activating Token Index: 52

Text #14

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0591. Min Act: -0.0001

Data Index: 1367897 (Python Code)

Max Activating Token Index: 418

Text #15

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0636. Min Act: -0.0001

Data Index: 332665 (C4 (Web Text))

Max Activating Token Index: 18

Text #16

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0652. Min Act: -0.0001

Data Index: 424976 (C4 (Web Text))

Max Activating Token Index: 887

Text #17

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0655. Min Act: -0.0001

Data Index: 840266 (C4 (Web Text))

Max Activating Token Index: 440

Text #18

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0626. Min Act: -0.0001

Data Index: 1226283 (C4 (Web Text))

Max Activating Token Index: 13

Text #19

Max Range: 0.1096. Min Range: -0.1096

Max Act: 0.0646. Min Act: -0.0001

Data Index: 1259349 (C4 (Web Text))

Max Activating Token Index: 171