Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1413 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.5131. Min Act: -0.0001

Data Index: 35069 (C4 (Web Text))

Max Activating Token Index: 39

Text #1

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.5001. Min Act: -0.0001

Data Index: 683634 (C4 (Web Text))

Max Activating Token Index: 800

Text #2

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.4549. Min Act: -0.0001

Data Index: 138185 (C4 (Web Text))

Max Activating Token Index: 564

Text #3

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.4355. Min Act: -0.0001

Data Index: 685529 (C4 (Web Text))

Max Activating Token Index: 170

Text #4

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.4057. Min Act: -0.0001

Data Index: 1077877 (C4 (Web Text))

Max Activating Token Index: 168

Text #5

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.4198. Min Act: -0.0001

Data Index: 932625 (C4 (Web Text))

Max Activating Token Index: 89

Text #6

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.4053. Min Act: -0.0001

Data Index: 821337 (C4 (Web Text))

Max Activating Token Index: 821

Text #7

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3951. Min Act: -0.0001

Data Index: 893101 (C4 (Web Text))

Max Activating Token Index: 297

Text #8

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.4050. Min Act: -0.0001

Data Index: 1379454 (Python Code)

Max Activating Token Index: 462

Text #9

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3658. Min Act: -0.0001

Data Index: 52036 (C4 (Web Text))

Max Activating Token Index: 890

Text #10

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.4042. Min Act: -0.0001

Data Index: 1283233 (C4 (Web Text))

Max Activating Token Index: 937

Text #11

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3992. Min Act: -0.0001

Data Index: 104518 (C4 (Web Text))

Max Activating Token Index: 843

Text #12

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3954. Min Act: -0.0001

Data Index: 447932 (C4 (Web Text))

Max Activating Token Index: 206

Text #13

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3636. Min Act: -0.0001

Data Index: 1498352 (Python Code)

Max Activating Token Index: 948

Text #14

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3983. Min Act: -0.0000

Data Index: 983940 (C4 (Web Text))

Max Activating Token Index: 539

Text #15

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3567. Min Act: -0.0000

Data Index: 381409 (C4 (Web Text))

Max Activating Token Index: 913

Text #16

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3741. Min Act: -0.0001

Data Index: 331523 (C4 (Web Text))

Max Activating Token Index: 34

Text #17

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3903. Min Act: -0.0001

Data Index: 350373 (C4 (Web Text))

Max Activating Token Index: 478

Text #18

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3694. Min Act: -0.0001

Data Index: 1381428 (Python Code)

Max Activating Token Index: 752

Text #19

Max Range: 0.5131. Min Range: -0.5131

Max Act: 0.3878. Min Act: -0.0001

Data Index: 1145521 (C4 (Web Text))

Max Activating Token Index: 221