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

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

Neuron 2564 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: 0.5715. Min Range: -0.5715

Max Act: 0.5715. Min Act: -0.0001

Data Index: 748126 (C4 (Web Text))

Max Activating Token Index: 578

Text #1

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.5475. Min Act: -0.0001

Data Index: 771464 (C4 (Web Text))

Max Activating Token Index: 272

Text #2

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.4867. Min Act: -0.0001

Data Index: 1205026 (C4 (Web Text))

Max Activating Token Index: 816

Text #3

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.4952. Min Act: -0.0001

Data Index: 890239 (C4 (Web Text))

Max Activating Token Index: 293

Text #4

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.4505. Min Act: -0.0001

Data Index: 1371416 (Python Code)

Max Activating Token Index: 275

Text #5

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.4638. Min Act: -0.0001

Data Index: 902584 (C4 (Web Text))

Max Activating Token Index: 63

Text #6

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.4303. Min Act: -0.0001

Data Index: 1316867 (C4 (Web Text))

Max Activating Token Index: 590

Text #7

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.4191. Min Act: -0.0001

Data Index: 967410 (C4 (Web Text))

Max Activating Token Index: 935

Text #8

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.4304. Min Act: -0.0001

Data Index: 1206013 (C4 (Web Text))

Max Activating Token Index: 353

Text #9

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.4027. Min Act: -0.0001

Data Index: 1085863 (C4 (Web Text))

Max Activating Token Index: 280

Text #10

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3921. Min Act: -0.0001

Data Index: 1372978 (Python Code)

Max Activating Token Index: 827

Text #11

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3845. Min Act: -0.0001

Data Index: 643921 (C4 (Web Text))

Max Activating Token Index: 214

Text #12

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3772. Min Act: -0.0001

Data Index: 730508 (C4 (Web Text))

Max Activating Token Index: 609

Text #13

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3834. Min Act: -0.0001

Data Index: 1600042 (Python Code)

Max Activating Token Index: 200

Text #14

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3843. Min Act: -0.0001

Data Index: 968231 (C4 (Web Text))

Max Activating Token Index: 939

Text #15

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3834. Min Act: -0.0001

Data Index: 646517 (C4 (Web Text))

Max Activating Token Index: 113

Text #16

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3821. Min Act: -0.0001

Data Index: 226607 (C4 (Web Text))

Max Activating Token Index: 464

Text #17

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3693. Min Act: -0.0001

Data Index: 546661 (C4 (Web Text))

Max Activating Token Index: 341

Text #18

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3753. Min Act: -0.0001

Data Index: 240179 (C4 (Web Text))

Max Activating Token Index: 285

Text #19

Max Range: 0.5715. Min Range: -0.5715

Max Act: 0.3658. Min Act: -0.0001

Data Index: 262107 (C4 (Web Text))

Max Activating Token Index: 662