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

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

Neuron 1378 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: 1.5535. Min Range: -1.5535

Max Act: 1.5535. Min Act: -0.0001

Data Index: 463807 (C4 (Web Text))

Max Activating Token Index: 61

Text #1

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.4231. Min Act: -0.0001

Data Index: 1003605 (C4 (Web Text))

Max Activating Token Index: 1005

Text #2

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1915. Min Act: -0.0001

Data Index: 771043 (C4 (Web Text))

Max Activating Token Index: 498

Text #3

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1471. Min Act: -0.0001

Data Index: 30615 (C4 (Web Text))

Max Activating Token Index: 183

Text #4

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1119. Min Act: -0.0001

Data Index: 983398 (C4 (Web Text))

Max Activating Token Index: 818

Text #5

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1780. Min Act: -0.0001

Data Index: 1383643 (Python Code)

Max Activating Token Index: 507

Text #6

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1651. Min Act: -0.0001

Data Index: 919419 (C4 (Web Text))

Max Activating Token Index: 548

Text #7

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1902. Min Act: -0.0001

Data Index: 1332502 (C4 (Web Text))

Max Activating Token Index: 88

Text #8

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1233. Min Act: -0.0001

Data Index: 1216096 (C4 (Web Text))

Max Activating Token Index: 1018

Text #9

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1128. Min Act: -0.0001

Data Index: 1290645 (C4 (Web Text))

Max Activating Token Index: 568

Text #10

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1324. Min Act: -0.0001

Data Index: 226027 (C4 (Web Text))

Max Activating Token Index: 229

Text #11

Max Range: 1.5535. Min Range: -1.5535

Max Act: 0.9063. Min Act: -0.0001

Data Index: 546953 (C4 (Web Text))

Max Activating Token Index: 547

Text #12

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1231. Min Act: -0.0001

Data Index: 36305 (C4 (Web Text))

Max Activating Token Index: 274

Text #13

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.0530. Min Act: -0.0001

Data Index: 1192143 (C4 (Web Text))

Max Activating Token Index: 3

Text #14

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.0537. Min Act: -0.0001

Data Index: 1570265 (Python Code)

Max Activating Token Index: 332

Text #15

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.0843. Min Act: -0.0001

Data Index: 39357 (C4 (Web Text))

Max Activating Token Index: 669

Text #16

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.1135. Min Act: -0.0001

Data Index: 1206086 (C4 (Web Text))

Max Activating Token Index: 409

Text #17

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.0883. Min Act: -0.0001

Data Index: 1480008 (Python Code)

Max Activating Token Index: 689

Text #18

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.0676. Min Act: -0.0001

Data Index: 357018 (C4 (Web Text))

Max Activating Token Index: 789

Text #19

Max Range: 1.5535. Min Range: -1.5535

Max Act: 1.0918. Min Act: -0.0001

Data Index: 338539 (C4 (Web Text))

Max Activating Token Index: 791