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

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

Neuron 449 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: 0.8404. Min Range: -0.8404

Max Act: 0.7966. Min Act: -0.0001

Data Index: 1164517 (C4 (Web Text))

Max Activating Token Index: 153

Text #1

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.7915. Min Act: -0.0001

Data Index: 899110 (C4 (Web Text))

Max Activating Token Index: 737

Text #2

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.8404. Min Act: -0.0001

Data Index: 1061725 (C4 (Web Text))

Max Activating Token Index: 747

Text #3

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.7938. Min Act: -0.0001

Data Index: 807460 (C4 (Web Text))

Max Activating Token Index: 726

Text #4

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6944. Min Act: -0.0001

Data Index: 977979 (C4 (Web Text))

Max Activating Token Index: 476

Text #5

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.7152. Min Act: -0.0001

Data Index: 617147 (C4 (Web Text))

Max Activating Token Index: 142

Text #6

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.7118. Min Act: -0.0001

Data Index: 13786 (C4 (Web Text))

Max Activating Token Index: 614

Text #7

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.7332. Min Act: -0.0001

Data Index: 810142 (C4 (Web Text))

Max Activating Token Index: 225

Text #8

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6900. Min Act: -0.0001

Data Index: 348130 (C4 (Web Text))

Max Activating Token Index: 993

Text #9

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.7090. Min Act: -0.0001

Data Index: 434850 (C4 (Web Text))

Max Activating Token Index: 266

Text #10

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6834. Min Act: -0.0001

Data Index: 457677 (C4 (Web Text))

Max Activating Token Index: 273

Text #11

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6798. Min Act: -0.0001

Data Index: 1177544 (C4 (Web Text))

Max Activating Token Index: 931

Text #12

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6861. Min Act: -0.0001

Data Index: 335453 (C4 (Web Text))

Max Activating Token Index: 740

Text #13

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6374. Min Act: -0.0001

Data Index: 1563539 (Python Code)

Max Activating Token Index: 11

Text #14

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6558. Min Act: -0.0001

Data Index: 827535 (C4 (Web Text))

Max Activating Token Index: 717

Text #15

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6733. Min Act: -0.0001

Data Index: 1190582 (C4 (Web Text))

Max Activating Token Index: 245

Text #16

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6695. Min Act: -0.0001

Data Index: 1058912 (C4 (Web Text))

Max Activating Token Index: 15

Text #17

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6507. Min Act: -0.0001

Data Index: 414726 (C4 (Web Text))

Max Activating Token Index: 587

Text #18

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6505. Min Act: -0.0001

Data Index: 1318534 (C4 (Web Text))

Max Activating Token Index: 397

Text #19

Max Range: 0.8404. Min Range: -0.8404

Max Act: 0.6478. Min Act: -0.0001

Data Index: 887592 (C4 (Web Text))

Max Activating Token Index: 717