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

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

Neuron 1534 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1265. Min Act: -0.0000

Data Index: 737374 (C4 (Web Text))

Max Activating Token Index: 289

Text #1

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1158. Min Act: -0.0001

Data Index: 1404513 (Python Code)

Max Activating Token Index: 279

Text #2

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1060. Min Act: -0.0000

Data Index: 1127745 (C4 (Web Text))

Max Activating Token Index: 209

Text #3

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1076. Min Act: -0.0000

Data Index: 566091 (C4 (Web Text))

Max Activating Token Index: 608

Text #4

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0999. Min Act: -0.0000

Data Index: 735278 (C4 (Web Text))

Max Activating Token Index: 198

Text #5

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1031. Min Act: -0.0000

Data Index: 905351 (C4 (Web Text))

Max Activating Token Index: 312

Text #6

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1026. Min Act: -0.0000

Data Index: 1183364 (C4 (Web Text))

Max Activating Token Index: 939

Text #7

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1055. Min Act: -0.0000

Data Index: 971768 (C4 (Web Text))

Max Activating Token Index: 245

Text #8

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0985. Min Act: -0.0001

Data Index: 7548 (C4 (Web Text))

Max Activating Token Index: 208

Text #9

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1009. Min Act: -0.0000

Data Index: 89511 (C4 (Web Text))

Max Activating Token Index: 359

Text #10

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1029. Min Act: -0.0000

Data Index: 39053 (C4 (Web Text))

Max Activating Token Index: 153

Text #11

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0979. Min Act: -0.0001

Data Index: 1416318 (Python Code)

Max Activating Token Index: 371

Text #12

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.1022. Min Act: -0.0000

Data Index: 1199598 (C4 (Web Text))

Max Activating Token Index: 742

Text #13

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0948. Min Act: -0.0000

Data Index: 599133 (C4 (Web Text))

Max Activating Token Index: 630

Text #14

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0932. Min Act: -0.0000

Data Index: 848579 (C4 (Web Text))

Max Activating Token Index: 858

Text #15

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0922. Min Act: -0.0000

Data Index: 845335 (C4 (Web Text))

Max Activating Token Index: 132

Text #16

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0947. Min Act: -0.0001

Data Index: 1455975 (Python Code)

Max Activating Token Index: 332

Text #17

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0872. Min Act: -0.0001

Data Index: 308088 (C4 (Web Text))

Max Activating Token Index: 24

Text #18

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0871. Min Act: -0.0000

Data Index: 499809 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.1265. Min Range: -0.1265

Max Act: 0.0863. Min Act: -0.0000

Data Index: 1330970 (C4 (Web Text))

Max Activating Token Index: 561