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

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

Neuron 2851 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: 3.8472. Min Range: -3.8472

Max Act: 3.8472. Min Act: -0.0001

Data Index: 381738 (C4 (Web Text))

Max Activating Token Index: 20

Text #1

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.7347. Min Act: -0.0001

Data Index: 1193912 (C4 (Web Text))

Max Activating Token Index: 615

Text #2

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5685. Min Act: -0.0001

Data Index: 1139639 (C4 (Web Text))

Max Activating Token Index: 167

Text #3

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5983. Min Act: -0.0001

Data Index: 1339157 (C4 (Web Text))

Max Activating Token Index: 102

Text #4

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.4594. Min Act: -0.0001

Data Index: 274828 (C4 (Web Text))

Max Activating Token Index: 43

Text #5

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5291. Min Act: -0.0001

Data Index: 235605 (C4 (Web Text))

Max Activating Token Index: 77

Text #6

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.6582. Min Act: -0.0001

Data Index: 1291826 (C4 (Web Text))

Max Activating Token Index: 692

Text #7

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5960. Min Act: -0.0001

Data Index: 747700 (C4 (Web Text))

Max Activating Token Index: 49

Text #8

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.3118. Min Act: -0.0001

Data Index: 477908 (C4 (Web Text))

Max Activating Token Index: 661

Text #9

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.2724. Min Act: -0.0001

Data Index: 282669 (C4 (Web Text))

Max Activating Token Index: 891

Text #10

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.4205. Min Act: -0.0001

Data Index: 273485 (C4 (Web Text))

Max Activating Token Index: 413

Text #11

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.4044. Min Act: -0.0001

Data Index: 239742 (C4 (Web Text))

Max Activating Token Index: 323

Text #12

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.4156. Min Act: -0.0001

Data Index: 187653 (C4 (Web Text))

Max Activating Token Index: 793

Text #13

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.3667. Min Act: -0.0001

Data Index: 158314 (C4 (Web Text))

Max Activating Token Index: 709

Text #14

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5251. Min Act: -0.0001

Data Index: 774940 (C4 (Web Text))

Max Activating Token Index: 511

Text #15

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.2715. Min Act: -0.0001

Data Index: 992200 (C4 (Web Text))

Max Activating Token Index: 815

Text #16

Max Range: 3.8472. Min Range: -3.8472

Max Act: 2.8722. Min Act: -0.0001

Data Index: 281770 (C4 (Web Text))

Max Activating Token Index: 560

Text #17

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.1133. Min Act: -0.0001

Data Index: 1245194 (C4 (Web Text))

Max Activating Token Index: 242

Text #18

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.0627. Min Act: -0.0001

Data Index: 149666 (C4 (Web Text))

Max Activating Token Index: 1006

Text #19

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.1724. Min Act: -0.0001

Data Index: 312217 (C4 (Web Text))

Max Activating Token Index: 201