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

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

Neuron 3548 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: 0.7761. Min Range: -0.7761

Max Act: 0.7761. Min Act: -0.0001

Data Index: 522963 (C4 (Web Text))

Max Activating Token Index: 585

Text #1

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.5371. Min Act: 0.0000

Data Index: 1654367 (Python Code)

Max Activating Token Index: 202

Text #2

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.5564. Min Act: -0.0001

Data Index: 1649786 (Python Code)

Max Activating Token Index: 449

Text #3

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.5078. Min Act: -0.0000

Data Index: 542954 (C4 (Web Text))

Max Activating Token Index: 682

Text #4

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.5545. Min Act: -0.0001

Data Index: 123816 (C4 (Web Text))

Max Activating Token Index: 767

Text #5

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.5294. Min Act: -0.0000

Data Index: 1491476 (Python Code)

Max Activating Token Index: 297

Text #6

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4881. Min Act: -0.0000

Data Index: 1302521 (C4 (Web Text))

Max Activating Token Index: 471

Text #7

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4857. Min Act: 0.0000

Data Index: 1525583 (Python Code)

Max Activating Token Index: 1011

Text #8

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4678. Min Act: 0.0000

Data Index: 1574578 (Python Code)

Max Activating Token Index: 316

Text #9

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4547. Min Act: -0.0000

Data Index: 1211428 (C4 (Web Text))

Max Activating Token Index: 81

Text #10

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4635. Min Act: -0.0000

Data Index: 242302 (C4 (Web Text))

Max Activating Token Index: 171

Text #11

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4506. Min Act: -0.0000

Data Index: 604534 (C4 (Web Text))

Max Activating Token Index: 288

Text #12

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4207. Min Act: -0.0000

Data Index: 537274 (C4 (Web Text))

Max Activating Token Index: 670

Text #13

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4164. Min Act: -0.0000

Data Index: 1602369 (Python Code)

Max Activating Token Index: 157

Text #14

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4178. Min Act: 0.0000

Data Index: 1433102 (Python Code)

Max Activating Token Index: 127

Text #15

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4414. Min Act: -0.0000

Data Index: 1582649 (Python Code)

Max Activating Token Index: 801

Text #16

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4084. Min Act: -0.0001

Data Index: 1163148 (C4 (Web Text))

Max Activating Token Index: 179

Text #17

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4338. Min Act: 0.0000

Data Index: 66276 (C4 (Web Text))

Max Activating Token Index: 600

Text #18

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.4017. Min Act: -0.0001

Data Index: 1628708 (Python Code)

Max Activating Token Index: 697

Text #19

Max Range: 0.7761. Min Range: -0.7761

Max Act: 0.3992. Min Act: -0.0000

Data Index: 1083317 (C4 (Web Text))

Max Activating Token Index: 578