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

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

Neuron 145 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.3027. Min Act: -0.0001

Data Index: 621514 (C4 (Web Text))

Max Activating Token Index: 820

Text #1

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.2318. Min Act: -0.0001

Data Index: 197168 (C4 (Web Text))

Max Activating Token Index: 596

Text #2

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.1294. Min Act: -0.0001

Data Index: 374656 (C4 (Web Text))

Max Activating Token Index: 862

Text #3

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.1750. Min Act: -0.0001

Data Index: 507712 (C4 (Web Text))

Max Activating Token Index: 521

Text #4

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.2269. Min Act: -0.0001

Data Index: 1173879 (C4 (Web Text))

Max Activating Token Index: 310

Text #5

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.2101. Min Act: -0.0001

Data Index: 1064289 (C4 (Web Text))

Max Activating Token Index: 816

Text #6

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.1617. Min Act: -0.0001

Data Index: 1252375 (C4 (Web Text))

Max Activating Token Index: 988

Text #7

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.1547. Min Act: -0.0001

Data Index: 969518 (C4 (Web Text))

Max Activating Token Index: 986

Text #8

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0620. Min Act: -0.0001

Data Index: 258226 (C4 (Web Text))

Max Activating Token Index: 645

Text #9

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0306. Min Act: -0.0001

Data Index: 1350543 (C4 (Web Text))

Max Activating Token Index: 805

Text #10

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0515. Min Act: -0.0001

Data Index: 523949 (C4 (Web Text))

Max Activating Token Index: 981

Text #11

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0323. Min Act: -0.0001

Data Index: 1275131 (C4 (Web Text))

Max Activating Token Index: 374

Text #12

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0478. Min Act: -0.0001

Data Index: 1039394 (C4 (Web Text))

Max Activating Token Index: 11

Text #13

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0105. Min Act: -0.0001

Data Index: 1133580 (C4 (Web Text))

Max Activating Token Index: 834

Text #14

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0397. Min Act: -0.0001

Data Index: 494015 (C4 (Web Text))

Max Activating Token Index: 28

Text #15

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0098. Min Act: -0.0001

Data Index: 1551283 (Python Code)

Max Activating Token Index: 207

Text #16

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0411. Min Act: -0.0001

Data Index: 863729 (C4 (Web Text))

Max Activating Token Index: 579

Text #17

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0674. Min Act: -0.0001

Data Index: 328571 (C4 (Web Text))

Max Activating Token Index: 760

Text #18

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0110. Min Act: -0.0001

Data Index: 803745 (C4 (Web Text))

Max Activating Token Index: 569

Text #19

Max Range: 1.3027. Min Range: -1.3027

Max Act: 1.0464. Min Act: -0.0001

Data Index: 370131 (C4 (Web Text))

Max Activating Token Index: 853