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

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

Neuron 2595 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: 1.2335. Min Range: -1.2335

Max Act: 1.2335. Min Act: -0.0001

Data Index: 1337636 (C4 (Web Text))

Max Activating Token Index: 876

Text #1

Max Range: 1.2335. Min Range: -1.2335

Max Act: 1.1117. Min Act: -0.0001

Data Index: 377034 (C4 (Web Text))

Max Activating Token Index: 486

Text #2

Max Range: 1.2335. Min Range: -1.2335

Max Act: 1.0726. Min Act: -0.0001

Data Index: 1443141 (Python Code)

Max Activating Token Index: 938

Text #3

Max Range: 1.2335. Min Range: -1.2335

Max Act: 1.0711. Min Act: -0.0001

Data Index: 1251227 (C4 (Web Text))

Max Activating Token Index: 594

Text #4

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.9438. Min Act: -0.0001

Data Index: 83652 (C4 (Web Text))

Max Activating Token Index: 500

Text #5

Max Range: 1.2335. Min Range: -1.2335

Max Act: 1.0231. Min Act: -0.0001

Data Index: 1632613 (Python Code)

Max Activating Token Index: 430

Text #6

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.9492. Min Act: -0.0001

Data Index: 576259 (C4 (Web Text))

Max Activating Token Index: 691

Text #7

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.8806. Min Act: -0.0001

Data Index: 339 (C4 (Web Text))

Max Activating Token Index: 901

Text #8

Max Range: 1.2335. Min Range: -1.2335

Max Act: 1.0198. Min Act: -0.0001

Data Index: 1039079 (C4 (Web Text))

Max Activating Token Index: 720

Text #9

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.9534. Min Act: -0.0001

Data Index: 1165697 (C4 (Web Text))

Max Activating Token Index: 299

Text #10

Max Range: 1.2335. Min Range: -1.2335

Max Act: 1.0439. Min Act: -0.0001

Data Index: 661361 (C4 (Web Text))

Max Activating Token Index: 301

Text #11

Max Range: 1.2335. Min Range: -1.2335

Max Act: 1.0101. Min Act: -0.0001

Data Index: 346396 (C4 (Web Text))

Max Activating Token Index: 823

Text #12

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.9965. Min Act: -0.0001

Data Index: 1067838 (C4 (Web Text))

Max Activating Token Index: 1021

Text #13

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.8903. Min Act: -0.0001

Data Index: 675359 (C4 (Web Text))

Max Activating Token Index: 674

Text #14

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.8806. Min Act: -0.0001

Data Index: 1464778 (Python Code)

Max Activating Token Index: 497

Text #15

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.9290. Min Act: -0.0001

Data Index: 1591484 (Python Code)

Max Activating Token Index: 481

Text #16

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.9579. Min Act: -0.0001

Data Index: 1086193 (C4 (Web Text))

Max Activating Token Index: 991

Text #17

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.9156. Min Act: -0.0001

Data Index: 1474013 (Python Code)

Max Activating Token Index: 557

Text #18

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.8988. Min Act: -0.0001

Data Index: 679932 (C4 (Web Text))

Max Activating Token Index: 597

Text #19

Max Range: 1.2335. Min Range: -1.2335

Max Act: 0.9577. Min Act: -0.0001

Data Index: 828949 (C4 (Web Text))

Max Activating Token Index: 330