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

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

Neuron 4339 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4953. Min Range: -1.4953

Max Act: 1.4953. Min Act: -0.0001

Data Index: 1337824 (C4 (Web Text))

Max Activating Token Index: 105

Text #1

Max Range: 1.4953. Min Range: -1.4953

Max Act: 1.4049. Min Act: -0.0001

Data Index: 282724 (C4 (Web Text))

Max Activating Token Index: 654

Text #2

Max Range: 1.4953. Min Range: -1.4953

Max Act: 1.3138. Min Act: -0.0001

Data Index: 330173 (C4 (Web Text))

Max Activating Token Index: 767

Text #3

Max Range: 1.4953. Min Range: -1.4953

Max Act: 1.2026. Min Act: -0.0001

Data Index: 829328 (C4 (Web Text))

Max Activating Token Index: 92

Text #4

Max Range: 1.4953. Min Range: -1.4953

Max Act: 1.0893. Min Act: -0.0001

Data Index: 407654 (C4 (Web Text))

Max Activating Token Index: 938

Text #5

Max Range: 1.4953. Min Range: -1.4953

Max Act: 1.0615. Min Act: -0.0001

Data Index: 1295398 (C4 (Web Text))

Max Activating Token Index: 980

Text #6

Max Range: 1.4953. Min Range: -1.4953

Max Act: 1.0127. Min Act: -0.0001

Data Index: 152870 (C4 (Web Text))

Max Activating Token Index: 206

Text #7

Max Range: 1.4953. Min Range: -1.4953

Max Act: 1.0633. Min Act: -0.0001

Data Index: 1375224 (Python Code)

Max Activating Token Index: 849

Text #8

Max Range: 1.4953. Min Range: -1.4953

Max Act: 1.0545. Min Act: -0.0001

Data Index: 423945 (C4 (Web Text))

Max Activating Token Index: 708

Text #9

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.9948. Min Act: -0.0001

Data Index: 1208816 (C4 (Web Text))

Max Activating Token Index: 181

Text #10

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.9884. Min Act: -0.0001

Data Index: 680081 (C4 (Web Text))

Max Activating Token Index: 149

Text #11

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.8994. Min Act: -0.0001

Data Index: 373907 (C4 (Web Text))

Max Activating Token Index: 587

Text #12

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.8720. Min Act: -0.0001

Data Index: 42972 (C4 (Web Text))

Max Activating Token Index: 676

Text #13

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.9249. Min Act: -0.0001

Data Index: 447361 (C4 (Web Text))

Max Activating Token Index: 474

Text #14

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.8535. Min Act: -0.0001

Data Index: 440887 (C4 (Web Text))

Max Activating Token Index: 751

Text #15

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.9175. Min Act: -0.0001

Data Index: 771166 (C4 (Web Text))

Max Activating Token Index: 999

Text #16

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.9508. Min Act: -0.0001

Data Index: 456036 (C4 (Web Text))

Max Activating Token Index: 390

Text #17

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.8004. Min Act: -0.0001

Data Index: 470996 (C4 (Web Text))

Max Activating Token Index: 552

Text #18

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.9527. Min Act: -0.0001

Data Index: 1092740 (C4 (Web Text))

Max Activating Token Index: 379

Text #19

Max Range: 1.4953. Min Range: -1.4953

Max Act: 0.8315. Min Act: -0.0001

Data Index: 778553 (C4 (Web Text))

Max Activating Token Index: 789