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

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

Neuron 2791 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.6755. Min Act: -0.0001

Data Index: 673043 (C4 (Web Text))

Max Activating Token Index: 932

Text #1

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.6740. Min Act: -0.0001

Data Index: 177891 (C4 (Web Text))

Max Activating Token Index: 467

Text #2

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.5597. Min Act: -0.0001

Data Index: 1254812 (C4 (Web Text))

Max Activating Token Index: 487

Text #3

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4957. Min Act: -0.0001

Data Index: 1113179 (C4 (Web Text))

Max Activating Token Index: 894

Text #4

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4747. Min Act: -0.0001

Data Index: 256283 (C4 (Web Text))

Max Activating Token Index: 254

Text #5

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4885. Min Act: -0.0001

Data Index: 986921 (C4 (Web Text))

Max Activating Token Index: 403

Text #6

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4765. Min Act: -0.0001

Data Index: 509852 (C4 (Web Text))

Max Activating Token Index: 893

Text #7

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4681. Min Act: -0.0001

Data Index: 724981 (C4 (Web Text))

Max Activating Token Index: 173

Text #8

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4634. Min Act: -0.0001

Data Index: 661079 (C4 (Web Text))

Max Activating Token Index: 277

Text #9

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4853. Min Act: -0.0001

Data Index: 304205 (C4 (Web Text))

Max Activating Token Index: 687

Text #10

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4261. Min Act: -0.0001

Data Index: 449928 (C4 (Web Text))

Max Activating Token Index: 87

Text #11

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4285. Min Act: -0.0001

Data Index: 1010665 (C4 (Web Text))

Max Activating Token Index: 339

Text #12

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4323. Min Act: -0.0001

Data Index: 1442003 (Python Code)

Max Activating Token Index: 378

Text #13

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4577. Min Act: -0.0001

Data Index: 342794 (C4 (Web Text))

Max Activating Token Index: 118

Text #14

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4411. Min Act: -0.0001

Data Index: 1626205 (Python Code)

Max Activating Token Index: 386

Text #15

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4207. Min Act: -0.0001

Data Index: 1538108 (Python Code)

Max Activating Token Index: 1016

Text #16

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4308. Min Act: -0.0001

Data Index: 1399099 (Python Code)

Max Activating Token Index: 400

Text #17

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4170. Min Act: -0.0001

Data Index: 1141709 (C4 (Web Text))

Max Activating Token Index: 56

Text #18

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4428. Min Act: -0.0001

Data Index: 228997 (C4 (Web Text))

Max Activating Token Index: 360

Text #19

Max Range: 0.6755. Min Range: -0.6755

Max Act: 0.4410. Min Act: -0.0001

Data Index: 631754 (C4 (Web Text))

Max Activating Token Index: 926