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

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

Neuron 1404 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.8985. Min Act: -0.0001

Data Index: 799446 (C4 (Web Text))

Max Activating Token Index: 245

Text #1

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.8266. Min Act: -0.0001

Data Index: 583421 (C4 (Web Text))

Max Activating Token Index: 323

Text #2

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.8129. Min Act: -0.0001

Data Index: 979378 (C4 (Web Text))

Max Activating Token Index: 989

Text #3

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.6050. Min Act: -0.0001

Data Index: 279642 (C4 (Web Text))

Max Activating Token Index: 141

Text #4

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.5104. Min Act: -0.0001

Data Index: 480560 (C4 (Web Text))

Max Activating Token Index: 357

Text #5

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.4994. Min Act: -0.0002

Data Index: 1234781 (C4 (Web Text))

Max Activating Token Index: 982

Text #6

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.3239. Min Act: -0.0001

Data Index: 549223 (C4 (Web Text))

Max Activating Token Index: 519

Text #7

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.6145. Min Act: -0.0001

Data Index: 1022100 (C4 (Web Text))

Max Activating Token Index: 320

Text #8

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.5288. Min Act: -0.0001

Data Index: 774515 (C4 (Web Text))

Max Activating Token Index: 710

Text #9

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.5141. Min Act: -0.0001

Data Index: 111544 (C4 (Web Text))

Max Activating Token Index: 89

Text #10

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.3654. Min Act: -0.0001

Data Index: 761963 (C4 (Web Text))

Max Activating Token Index: 813

Text #11

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.0738. Min Act: -0.0001

Data Index: 739021 (C4 (Web Text))

Max Activating Token Index: 448

Text #12

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.4488. Min Act: -0.0001

Data Index: 212258 (C4 (Web Text))

Max Activating Token Index: 516

Text #13

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.1268. Min Act: -0.0002

Data Index: 315320 (C4 (Web Text))

Max Activating Token Index: 186

Text #14

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.2710. Min Act: -0.0001

Data Index: 1080302 (C4 (Web Text))

Max Activating Token Index: 991

Text #15

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.2524. Min Act: -0.0001

Data Index: 902440 (C4 (Web Text))

Max Activating Token Index: 322

Text #16

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.2058. Min Act: -0.0001

Data Index: 599410 (C4 (Web Text))

Max Activating Token Index: 558

Text #17

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.0523. Min Act: -0.0002

Data Index: 282112 (C4 (Web Text))

Max Activating Token Index: 485

Text #18

Max Range: 3.8985. Min Range: -3.8985

Max Act: 2.9144. Min Act: -0.0001

Data Index: 1436030 (Python Code)

Max Activating Token Index: 866

Text #19

Max Range: 3.8985. Min Range: -3.8985

Max Act: 3.1675. Min Act: -0.0001

Data Index: 842079 (C4 (Web Text))

Max Activating Token Index: 625