Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 757 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.4025. Min Act: -0.0000

Data Index: 309745 (C4 (Web Text))

Max Activating Token Index: 373

Text #1

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.3118. Min Act: -0.0000

Data Index: 492823 (C4 (Web Text))

Max Activating Token Index: 479

Text #2

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.3119. Min Act: -0.0000

Data Index: 670735 (C4 (Web Text))

Max Activating Token Index: 993

Text #3

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.3023. Min Act: -0.0000

Data Index: 1450710 (Python Code)

Max Activating Token Index: 529

Text #4

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2840. Min Act: -0.0000

Data Index: 587718 (C4 (Web Text))

Max Activating Token Index: 117

Text #5

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2830. Min Act: -0.0000

Data Index: 841357 (C4 (Web Text))

Max Activating Token Index: 838

Text #6

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.3002. Min Act: -0.0000

Data Index: 1056648 (C4 (Web Text))

Max Activating Token Index: 964

Text #7

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2956. Min Act: -0.0000

Data Index: 1148963 (C4 (Web Text))

Max Activating Token Index: 378

Text #8

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2899. Min Act: -0.0000

Data Index: 1124358 (C4 (Web Text))

Max Activating Token Index: 364

Text #9

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2840. Min Act: -0.0000

Data Index: 764198 (C4 (Web Text))

Max Activating Token Index: 541

Text #10

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2807. Min Act: -0.0000

Data Index: 793169 (C4 (Web Text))

Max Activating Token Index: 218

Text #11

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2781. Min Act: -0.0000

Data Index: 31565 (C4 (Web Text))

Max Activating Token Index: 283

Text #12

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2669. Min Act: -0.0000

Data Index: 1390152 (Python Code)

Max Activating Token Index: 12

Text #13

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2742. Min Act: -0.0000

Data Index: 515932 (C4 (Web Text))

Max Activating Token Index: 541

Text #14

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2478. Min Act: -0.0000

Data Index: 1486744 (Python Code)

Max Activating Token Index: 630

Text #15

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2441. Min Act: -0.0000

Data Index: 921604 (C4 (Web Text))

Max Activating Token Index: 580

Text #16

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2599. Min Act: -0.0000

Data Index: 500449 (C4 (Web Text))

Max Activating Token Index: 615

Text #17

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2780. Min Act: -0.0000

Data Index: 1649595 (Python Code)

Max Activating Token Index: 242

Text #18

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2600. Min Act: -0.0000

Data Index: 967496 (C4 (Web Text))

Max Activating Token Index: 804

Text #19

Max Range: 0.4025. Min Range: -0.4025

Max Act: 0.2431. Min Act: -0.0000

Data Index: 1381169 (Python Code)

Max Activating Token Index: 843